Diffusion models are prone to exactly reproduce images from the training data. This exact reproduction of the training data is concerning as it can lead to copyright infringement and/or leakage of privacy-sensitive information. In this paper, we present a novel perspective on the memorization phenomenon and propose a simple yet effective approach to mitigate it. We argue that memorization occurs because of an attraction basin in the denoising process which steers the diffusion trajectory towards a memorized image. However, this can be mitigated by guiding the diffusion trajectory away from the attraction basin by not applying classifier-free guidance until an ideal transition point occurs from which classifier-free guidance is applied. This leads to the generation of non-memorized images that are high in image quality and well-aligned with the conditioning mechanism. To further improve on this, we present a new guidance technique, opposite guidance, that escapes the attraction basin sooner in the denoising process. We demonstrate the existence of attraction basins in various scenarios in which memorization occurs, and we show that our proposed approach successfully mitigates memorization. Our codebase is publicly available at https://github.com/SonyResearch/mitigating_memorization.

Generative models can reconstruct face images from encoded representations (templates) bearing remarkable likeness to the original face, raising security and privacy concerns. We present FACECLOAK, a neural network framework that protects face templates by generating smart, renewable binary cloaks. Our method proactively thwarts inversion attacks by cloaking face templates with unique disruptors synthesized from a single face template on the fly while provably retaining biometric utility and unlinkability. Our cloaked templates can suppress sensitive attributes while generalizing to novel feature extraction schemes and outperform leading baselines in terms of biometric matching and resiliency to reconstruction attacks. FACECLOAK-based matching is extremely fast (inference time =0.28 ms) and light (0.57 MB). We have released our code for reproducible research.

Deep generative models require large amounts of training data. This often poses a problem as the collection of datasets can be expensive and difficult, in particular datasets that are representative of the appropriate underlying distribution (e.g. demographic). This introduces biases in datasets which are further propagated in the models. We present an approach to construct an unbiased generative adversarial network (GAN) from an existing biased GAN by rebalancing the model distribution. We do so by generating balanced data from an existing imbalanced deep generative model using an evolutionary algorithm and then using this data to train a balanced generative model. Additionally, we propose a bias mitigation loss function that minimizes the deviation of the learned class distribution from being equiprobable. We show results for the StyleGAN2 models while training on the Flickr Faces High Quality (FFHQ) dataset for racial fairness and see that the proposed approach improves on the fairness metric by almost 5 times, whilst maintaining image quality. We further validate our approach by applying it to an imbalanced CIFAR10 dataset where we show that we can obtain comparable fairness and image quality as when training on a balanced CIFAR10 dataset which is also twice as large. Lastly, we argue that the traditionally used image quality metrics such as Frechet inception distance (FID) are unsuitable for scenarios where the class distributions are imbalanced and a balanced reference set is not available.

Radio remains a pervasive medium for mass information dissemination, with AM/FM stations reaching more Americans than either smartphone-based social networking or live television.Increasingly, radio broadcasts are also streamed online and accessed over the Internet.We present WavePulse, a framework that records, documents, and analyzes radio content in real-time.While our framework is generally applicable, we showcase the efficacy of WavePulse in a collaborative project with a team of political scientists focusing on the 2024 Presidential Election.We use WavePulse to monitor livestreams of 396 news radio stations over a period of three months, processing close to 500,000 hours of audio streams.These streams were converted into time-stamped, diarized transcripts and analyzed to answer key political science questions at both the national and state levels.Our analysis revealed how local issues interacted with national trends, providing insights into information flow.Our results demonstrate WavePulse's efficacy in capturing and analyzing content from radio livestreams sourced from the Web.Code and dataset can be accessed at https://wave-pulse.io

With the rise of AI-enabled Real-Time Deepfakes (RTDFs), the integrity of online video interactions has become a growing concern. RTDFs have now made it feasible to replace an imposter's face with their victim in live video interactions. Such advancement in deepfakes also coaxes detection to rise to the same standard. However, existing deepfake detection techniques are asynchronous and hence ill-suited for RTDFs. To bridge this gap, we propose a challenge-response approach that establishes authenticity in live settings. We focus on talking-head style video interaction and present a taxonomy of challenges that specifically target inherent limitations of RTDF generation pipelines. We evaluate representative examples from the taxonomy by collecting a unique dataset comprising eight challenges, which consistently and visibly degrades the quality of state-of-the-art deepfake generators. These results are corroborated both by humans and a new automated scoring function, leading to 88.6% and 80.1% AUC, respectively. The findings under-score the promising potential of challenge-response systems for explainable and scalable real-time deepfake detection in practical scenarios. We provide access to data and code at https://github.com/mittalgovind/GOTCHA-Deepfakes.

Through a large-scale study over diverse face images, we show that facial attribute editing using modern generative AI models can severely degrade automated face recognition systems. This degradation persists even with generative models that include additional identity-based loss function. To mitigate this issue, we propose two novel techniques for local and global attribute editing. We empirically ablate twenty-six facial semantic, demographic and expression-based attributes that have been edited using state-of-the-art generative models, and evaluate them using ArcFace and AdaFace matchers on CelebA, CelebAMaskHQ and LFW datasets. Finally, we use LLaVA, an emerging visual question-answering framework for attribute prediction to validate our editing techniques. Our methods outperform the current state-of-the-art at facial editing (BLIP, InstantID) while retaining identity by a significant extent. Our code is available at https://github.com/sudban3089/MIAFR.git.

The performance of automated face recognition systems is inevitably impacted by the facial aging process. However, high quality datasets of individuals collected over several years are typically small in scale. In this work, we propose, train, and validate the use of latent text-to-image diffusion models for synthetically aging and de-aging face images. Our models succeed with few-shot training, and have the added benefit of being controllable via intuitive textual prompting. We observe high degrees of visual realism in the generated images while maintaining biometric fidelity measured by commonly used metrics. We evaluate our method on two benchmark datasets (CelebA and AgeDB) and observe significant reduction (~ 44%) in the False Non-Match Rate compared to existing state-of the-art baselines.

Facial recognition systems have made significant strides thanks to data-heavy deep learning models, but these models rely on large privacy-sensitive datasets. Further, many of these datasets lack diversity in terms of ethnicity and demographics, which can lead to biased models that can have serious societal and security implications. To address these issues, we propose a methodology that leverages the biased generative model StyleGAN2 to create demographically diverse images of synthetic individuals. The synthetic dataset is created using a novel evolutionary search algorithm that targets specific demographic groups. By training face recognition models with the resulting balanced dataset containing 50,000 identities per race (13.5 million images in total), we can improve their performance and minimize biases that might have been present in a model trained on a real dataset.

In this paper, we propose dictionary attacks against speaker verification-a novel attack vector that aims to match a large fraction of speaker population by chance. We introduce a generic formulation of the attack that can be used with various speech representations and threat models. The attacker uses adversarial optimization to maximize raw similarity of speaker embeddings between a seed speech sample and a proxy population. The resulting master voice successfully matches a non-trivial fraction of people in an unknown population. Adversarial waveforms obtained with our approach can match on average 69% of females and 38% of males enrolled in the target system at a strict decision threshold calibrated to yield false alarm rate of 1%. By using the attack with a black-box voice cloning system, we obtain master voices that are effective in the most challenging conditions and transferable between speaker encoders. We also show that, combined with multiple attempts, this attack opens even more to serious issues on the security of these systems.

Due to a constant imbalance between demand and supply of power, ENF (Electric Network Frequency) fluctuates around a nominal value of 50 or 60 Hz. These variations in ENF cause the luminance intensity of a mains-powered light source, having no AC/DC converter inside, also to fluctuate. As a result, a video of a scene illuminated by a mains-powered light source can be used to estimate these fluctuations. As a consequence, the ENF signal within the time period when the video was captured can be estimated. This work explores the effects of frame rate harmonics that emerge when a rolling shutter based approach is used for ENF estimation from videos captured using CMOS cameras. These harmonics are a problem, especially for videos whose frame rate is a divisor of the nominal ENF because the frame rate harmonics and the ENF harmonics overlap. It is discovered that a key reason for the presence of the harmonics is the inverse square law of light that results in some repeating patterns of luminance variation across frames. This paper presents an analysis of the effect of the inverse square law of light on ENF estimation. A technique for refined ENF-related luminance signal estimation is proposed that attenuates these frame rate harmonics. This enables more accurate ENF estimates. The work also proposes an approach to estimate ENF-related luminance waveform cycles within each video frame, and a method to compute the confidence score for the estimated cycles. It provides insight into the reliability of the extracted ENF signal from a video, in the sense of its usefulness for ENF forensics, and consequently for ENF detection, which is an important precursor to ENF-based video forensics.

Face swapping technology used to create “Deepfakes” has advanced significantly over the past few years and now enables us to create realistic facial manipulations. Current deep learning algorithms to detect deepfakes have shown promising results, however, they require large amounts of training data, and as we show they are biased towards a particular ethnicity. We propose a deepfake detection methodology that eliminates the need for any real data by making use of synthetically generated data using Style-GAN3. This not only performs at par with the traditional training methodology of using real data but it shows better generalization capabilities when finetuned with a small amount of real data. Furthermore, this also reduces biases created by facial image datasets that might have sparse data from particular ethnicities. To promote reproducibility the code base has been made publicly available11https://github.com/anubhav1997/youneednodataset

Misinformation has developed into a critical societal threat that can lead to disastrous societal consequences. Although fact-checking plays a key role in combating misinformation, relatively little research has empirically investigated work practices of professional fact-checkers. To address this gap, we conducted semi-structured interviews with 21 fact-checkers from 19 countries. The participants reported being inundated with information that needs filtering and prioritizing prior to fact-checking. The interviews surfaced a pipeline of practices fragmented across disparate tools that lack integration. Importantly, fact-checkers lack effective mechanisms for disseminating the outcomes of their efforts which prevents their work from fully achieving its potential impact. We found that the largely manual and labor intensive nature of current fact-checking practices is a barrier to scale. We apply these findings to propose a number of suggestions that can improve the effectiveness, efficiency, scale, and reach of fact-checking work and its outcomes.

Electric Network Frequency (ENF) continuously fluctuates around a nominal value (50/60 Hz) due to a persistent imbalance between supplied and demanded power. In certain circumstances, ENF gets intrinsically embedded into audio and video recordings and can be extracted from these recordings. Consequently, ENF can be used in a number of media forensic applications, such as verifying the time of recording of the media. In this work, a robust media time-stamping approach is proposed for media whose ENF content is relatively contaminated. It essentially entails two procedures: first, detecting all useful, i.e., considerably accurate, samples of an estimated ENF signal, and then applying an adapted normalized cross-correlation process that is designed for exploiting just the selected ENF portions based on a binary mask of the identified accurate samples. Experimental results show that the proposed approach provides significantly increased performance.

Analysis of imaging sensors is one of the most reliable photo forensic techniques, but it is increasingly challenged by complex image processing in modern cameras. The underlying photo response non-uniformity (PRNU) is distilled into a static sensor fingerprint unique for each device. This makes it easy to estimate and spoof and limits its reliability in face of sophisticated attackers. We propose to exploit computational capabilities of emerging intelligent vision sensors to design next-generation computational sensor fingerprints. Such sensors allow for running neural network inference directly on raw pixels, which enables end-to-end optimization of the entire photo acquisition and distribution pipeline. Control over fingerprint generation allows for adaptation to various requirements and threat models. In this study we provide a detailed assessment of security properties and evaluate two approaches to prevent spoofing: fingerprint generation based on local image content and adversarial training. We found that adversarial training is currently impractical, but content fingerprints deliver good performance in the considered cross-domain (RAW-RGB) setting and could provide robust best-effort protection against photo manipulation. Moreover, computational fingerprints can alleviate other limitations of PRNU, e.g., its limited reliability for dark/texture content and expensive fingerprint storage that hinders scalability. To enable this line of work, we developed a novel open-source and high-fidelity simulation environment for modeling photo acquisition and distribution pipelines (https://github.com/pkorus/neural-imaging).

Photo Response Non-Uniformity (PRNU) noise obtained from images or videos is used as a camera fingerprint to attribute visual objects captured by a camera. The PRNU-based source attribution method, however, fails when there is misalignment between the fingerprint and the query object. One example of such a misalignment, which has been overlooked in the field, is caused by the in-camera resizing technique that a video may have been subjected to. This paper investigates the attribution of visual media in the context of matching a video query object to an image fingerprint or vice versa. Specifically this paper focuses on improving camera attribution performance by taking into account the effects of binning, a commonly used in-camera resizing technique applied to video. We experimentally show that the True Positive Rate (TPR) obtained when binning is considered is approximately 3% higher.

Can we leverage high-resolution information without the unsustainable quadratic complexity to input scale? We propose Traversal Network (TNet), a novel multi-scale hard-attention architecture, which traverses image scale-space in a top-down fashion, visiting only the most informative image regions along the way. TNet offers an adjustable trade-off between accuracy and complexity, by changing the number of attended image locations. We compare our model against hard-attention baselines on ImageNet, achieving higher accuracy with less resources (FLOPs, processing time and memory). We further test our model on fMoW dataset, where we process satellite images of size up to $896 \times 896$ px, getting up to $2.5$x faster processing compared to baselines operating on the same resolution, while achieving higher accuracy as well. TNet is modular, meaning that most classification models could be adopted as its backbone for feature extraction, making the reported performance gains orthogonal to benefits offered by existing optimized deep models. Finally, hard-attention guarantees a degree of interpretability to our model's predictions, without any extra cost beyond inference.

Cybersecurity suffers from an oversaturation of centralized, hierarchical systems and a lack of exploration in the area of horizontal security, or security techniques and technologies which utilize democratic participation for security decision-making. Because of this, many horizontally governed organizations such as activist groups, worker cooperatives, trade unions, not-for-profit associations, and others are not represented in current cybersecurity solutions, and are forced to adopt hierarchical solutions to cybersecurity problems. This causes power dynamic mismatches that lead to cybersecurity and organizational operations failures. In this work we introduce COLBAC, a collective based access control system aimed at addressing this lack. COLBAC uses democratically authorized capability tokens to express access control policies. It allows for a flexible and dynamic degree of horizontality to meet the needs of different horizontally governed organizations. After introducing COLBAC, we finish with a discussion on future work needed to realize more horizontal security techniques, tools, and technologies.

The necessity for a steady STEM workforce has prompted academia to develop strategies to encourage people of diverse backgrounds to enter the STEM fields. A bridge program, also known as a conversion program, offers alternative pathways for individuals who have no prior computing education to receive the education that can help in developing their careers or acquiring a graduate-level degree in the computer science fields. This mixed-methods study consisted of two parts. First, an online post-baccalaureate bridge program was evaluated, with a focus on students’ performance. Factors for analysis included gender, prior major, and the length of the program, any or all of which might play a role in students’ unsuccessful attempts to complete the program. The results indicated that female students have a higher tendency to not complete the program. However, female students who completed the program and enrolled in a graduate school have as much potential to do well in the MS program as their male cohorts do. The second part of the study comprised a survey of students who completed or did not complete the program and interviews with women students. Grounded in self-determination theory (SDT), the results showed that strategies are needed to enhance women students’ perceived competence and relatedness in the program.

We present FiFTy, a modern file-type identification tool for memory forensics and data carving. In contrast to previous approaches based on hand-crafted features, we design a compact neural network architecture, which uses a trainable embedding space. Our approach dispenses with the explicit feature extraction which has been a bottleneck in legacy systems. We evaluate the proposed method on a novel dataset with 75 filetypes - the most diverse and balanced dataset reported to date. FiFTy consistently outperforms all baselines in terms of speed, accuracy and individual misclassification rates. We achieved an average accuracy of 77.5% with processing speed of ≈38 sec/GB, which is better and more than an order of magnitude faster than the previous state-of-the-art tool - Sceadan (69% at 9 min/GB). Our tool and the corresponding dataset is open-source.

Fact checking by professionals is viewed as a vital defense in the fight against misinformation. While fact checking is important and its impact has been significant, fact checks could have limited visibility and may not reach the intended audience, such as those deeply embedded in polarized communities. Concerned citizens (i.e., the crowd), who are users of the platforms where misinformation appears, can play a crucial role in disseminating fact-checking information and in countering the spread of misinformation. To explore if this is the case, we conduct a data-driven study of misinformation on the Twitter platform, focusing on tweets related to the COVID-19 pandemic, analyzing the spread of misinformation, professional fact checks, and the crowds response to popular misleading claims about COVID-19.In this work, we curate a dataset of false claims and statements that seek to challenge or refute them. We train a classifier to create a novel dataset of 155,468 COVID-19-related tweets, containing 33,237 false claims and 33,413 refuting arguments. Our findings show that professional fact-checking tweets have limited volume and reach. In contrast, we observe that the surge in misinformation tweets results in a quick response and a corresponding increase in tweets that refute such misinformation. More importantly, we find contrasting differences in the way the crowd refutes tweets, some tweets appear to be opinions, while others contain concrete evidence, such as a link to a reputed source. Our work provides insights into how misinformation is organically countered in social platforms by some of their users and the role they play in amplifying professional fact checks. These insights could lead to development of tools and mechanisms that can empower concerned citizens in combating misinformation. The code and data can be found in this link.1

In recent years, social media services have been leveraged to spread fake news stories. Helping people spot fake stories by marking them with credibility indicators could dissuade them from sharing such stories, thus reducing their amplification. We carried out an online study (N = 1,512) to explore the impact of four types of credibility indicators on people's intent to share news headlines with their friends on social media. We confirmed that credibility indicators can indeed decrease the propensity to share fake news. However, the impact of the indicators varied, with fact checking services being the most effective. We further found notable differences in responses to the indicators based on demographic and personal characteristics and social media usage frequency. Our findings have important implications for curbing the spread of misinformation via social media platforms.

Photo Response Non-Uniformity (PRNU) based camera attribution is an effective method to determine the source camera of a visual object (an image or a video). To apply this method, images or videos need to be obtained from a camera to create a “camera fingerprint” which then can be compared against the PRNU of the query media whose origin is under question. The fingerprint extraction process can be time consuming when a large number of video frames or images have to be denoised. This may need to be done when the individual images have been subjected to high compression or other geometric processing such as video stabilization. This paper investigates a simple, yet effective and efficient technique to create a camera fingerprint when so many still images need to be denoised. The technique utilizes Spatial Domain Averaged (SDA) frames. An SDA-frame is the arithmetic mean of multiple still images. When it is used for fingerprint extraction, the number of denoising operations can be significantly decreased with little or no performance loss. Experimental results show that the proposed method can work more than 50 times faster than conventional methods while providing similar matching results.

Forensic analysis of digital photo provenance relies on intrinsic traces left in the photograph at the time of its acquisition. Such analysis becomes unreliable after heavy post-processing, such as down-sampling and re-compression applied upon distribution in the Web. This paper explores end-to-end optimization of the entire image acquisition and distribution workflow to facilitate reliable forensic analysis at the end of the distribution channel. We demonstrate that neural imaging pipelines can be trained to replace the internals of digital cameras, and jointly optimized for high-fidelity photo development and reliable provenance analysis. In our experiments, the proposed approach increased image manipulation detection accuracy from 45% to over 90%. The findings encourage further research towards building more reliable imaging pipelines with explicit provenance-guaranteeing properties.

ENF (Electric Network Frequency) oscillates around a nominal value (50/60 Hz) due to imbalance between consumed and generated power. The intensity of a light source powered by mains electricity varies depending on the ENF fluctuations. These fluctuations can be extracted from videos recorded in the presence of mains-powered source illumination. This work investigates how the quality of the ENF signal estimated from video is affected by different light source illumination, compression ratios, and by social media encoding. Also explored is the effect of the length of the ENF ground-truth database on time of recording detection and verification.

In this paper, we assess vulnerability of speaker verification systems to dictionary attacks. We seek master voices, i.e., adversarial utterances optimized to match against a large number of users by pure chance. First, we perform menagerie analysis to identify utterances which intrinsically hold this property. Then, we propose an adversarial optimization approach for generating master voices synthetically. Our experiments show that, even in the most secure configuration, on average, a master voice can match approx. 20% of females and 10% of males without any knowledge about the population. We demonstrate that dictionary attacks should be considered as a feasible threat model for sensitive and high-stakes deployments of speaker verification.

In this paper, we address the problem of forensic photo carving, which serves as one of the key sources of digital evidence in modern law enforcement. We propose efficient algorithms for assembling meaningful photographs from orphaned photofragments, carved without access to file headers, meta-data, or compression settings. The addressed problem raises a novel variant of a jigsaw puzzle with an unknown number of mixed images, missing pieces, and severe brightness and colorization artifacts. We construct an efficient compatibility metric for matching puzzle pieces and a corresponding image stitching procedure which allows us to mitigate these artifacts. To facilitate photo assembly, we perform a forensic analysis of the fragments to provide clues about their location within the frame of the imaging sensor. The proposed algorithm formulates the assembly problem as finding non-overlapping sets in an interval graph spanned over the input fragments. The algorithm exhibits lower computational complexity compared with a popular puzzle-solving approach based on minimal spanning trees.

Electric network frequency (ENF) is a time-varying signal of the frequency of mains electricity in a power grid. It continuously fluctuates around a nominal value (50/60 Hz) due to changes in the supply and demand of power over time. Depending on these ENF variations, the luminous intensity of a mains-powered light source also fluctuates. These fluctuations in luminance can be captured by video recordings. Accordingly, the ENF can be estimated from such videos by the analysis of steady content in the video scene. When videos are captured by using a rolling shutter sampling mechanism, as is done mostly with CMOS cameras, there is an idle period between successive frames. Consequently, a number of illumination samples of the scene are effectively lost due to the idle period. These missing samples affect the ENF estimation, in the sense of the frequency shift caused and the power attenuation that results. This paper develops an analytical model for videos captured using a rolling shutter mechanism. This model illustrates how the frequency of the main ENF harmonic varies depending on the idle period length, and how the power of the captured ENF attenuates as idle period increases. Based on this, a novel idle period estimation method for potential use in camera forensics that is able to operate independently of video frame rate is proposed. Finally, a novel time-of-recording verification approach based on the use of multiple ENF components, idle period assumptions, and the interpolation of missing ENF samples is also proposed.

The unavailability of a pornographic image database has been an impediment for automated detection of child pornographic images. Beyond data confidentiality and privacy issues, even mere possession of such images is illegal in many countries. In this paper, these issues are addressed for skin tone detection, which is an essential component for filtering pornographic images. A pornographic image is encrypted using order preserving encryption, randomization, and permutation. Skin pixels are detected from the encrypted image in the encrypted domain without revealing the image content. Experiments and analysis show that the proposed scheme has low overhead and no degradation in detection accuracy.

The strength of an anonymity system depends on the number of users. Therefore, User eXperience (UX) and usability of these systems is of critical importance for boosting adoption and use. To this end, we carried out a study with 19 non-expert participants to investigate how users experience routine Web browsing via the Tor Browser, focusing particularly on encountered problems and frustrations. Using a mixed-methods quantitative and qualitative approach to study one week of naturalistic use of the Tor Browser, we uncovered a variety of UX issues, such as broken Web sites, latency, lack of common browsing conveniences, differential treatment of Tor traffic, incorrect geolocation, operational opacity, etc. We applied this insight to suggest a number of UX improvements that could mitigate the issues and reduce user frustration when using the Tor Browser.

PRNU (Photo Response Non-Uniformity)-based camera fingerprints are useful for identifying the source camera of an anonymous image. As the query image has to be correlated with each candidate camera fingerprint, one key concern of this approach is the high run time overhead when using a large camera database. Clever techniques have been proposed to reduce the computation and I/O time either by reducing the size of the fingerprint or by group testing where multiple candidate fingerprints can be eliminated by a single correlation operation. However, these techniques assume that the query images have not been scaled or cropped. In practice this may often not be the case, especially when query images are taken from social media sites. This paper presents a simple scaling-based approach for source camera identification when the query image is of full resolution or if it is cropped from an unknown location (of the original image). The proposed approach can also be easily combined with other known approaches for PRNU matching of scaled images. Experiments using 250 cameras showed that the run time overhead can decrease by a factor 13 when the query is a cropped image.

Many network intrusion detection systems use byte sequences to detect lateral movements that exploit remote vulnerabilities. Attackers bypass such detection by stealing valid credentials and using them to transmit from one computer to another without creating abnormal network traffic. We call this method Credential-based Lateral Movement. To detect this type of lateral movement, we develop the concept of a Network Login Structure that specifies normal logins within a given network. Our method models a network login structure by automatically extracting a collection of login patterns by using a variation of the market-basket algorithm. We then employ an anomaly detection approach to detect malicious logins that are inconsistent with the enterprise network's login structure. Evaluations show that the proposed method is able to detect malicious logins in a real setting. In a simulated attack, our system was able to detect 82% of malicious logins, with a 0.3% false positive rate. We used a real dataset of millions of logins over the course of five months within a global financial company for evaluation of this work.

Given a query image or video, or a known camera fingerprint, there is a lack of capabilities for fast identification of media, from a large repository of images and videos, that match the query fingerprint. This work introduces a new approach that improves the computation efficiency of pairwise camera fingerprint matching and incorporates group testing to make the search more effective. More specifically, we jointly leverage the individual strengths of composite fingerprints and fingerprint digests in a novel manner and design two methods that are superior to existing approaches. The results show that under very high-performance requirements, where the probability of correct identification is close to one with a false-positive rate of zero, the proposed search methods are 2-8 times faster than the state-of-art search methods.

Electrical network frequency (ENF) instantaneously fluctuates around its nominal value (50/60 Hz) due to a continuous disparity between generated power and consumed power. Consequently, luminous intensity of a mains-powered light source varies depending on ENF fluctuations in the grid network. Variations in the luminance over time can be captured from video recordings and ENF can be estimated through content analysis of these recordings. In ENF-based video forensics, it is critical to check whether a given video file is appropriate for this type of analysis. That is, if ENF signal is not present in a given video, it would be useless to apply ENF-based forensic analysis. In this letter, an ENF signal presence detection method is introduced for videos. The proposed method is based on multiple ENF signal estimations from steady superpixels, i.e., pixels that are most likely uniform in color, brightness, and texture, and intra-class similarity of the estimated signals. Subsequently, consistency among these estimates is then used to determine the presence or absence of an ENF signal in a given video. The proposed technique can operate on video clips as short as 2 min and is independent of the camera sensor type, i.e., CCD or CMOS.

We address the problem of shoulder-surfing attacks on authentication schemes by proposing IllusionPIN (IPIN), a PIN-based authentication method that operates on touchscreen devices. IPIN uses the technique of hybrid images to blend two keypads with different digit orderings in such a way, that the user who is close to the device is seeing one keypad to enter her PIN, while the attacker who is looking at the device from a bigger distance is seeing only the other keypad. The user's keypad is shuffled in every authentication attempt, since the attacker may memorize the spatial arrangement of the pressed digits. To reason about the security of IPIN, we developed an algorithm which is based on human visual perception and estimates the minimum distance from which an observer is unable to interpret the keypad of the user. We tested our estimations with 84 simulated shoulder-surfing attacks from 21 different people. None of the attacks was successful against our estimations. In addition, we estimated the minimum distance from which a camera is unable to capture the visual information from the keypad of the user. Based on our analysis, it seems practically almost impossible for a surveillance camera to capture the PIN of a smartphone user when IPIN is in use.

This paper investigates the security of partial fingerprint-based authentication systems, especially when multiple fingerprints of a user are enrolled. A number of consumer electronic devices, such as smartphones, are beginning to incorporate fingerprint sensors for user authentication. The sensors embedded in these devices are generally small and the resulting images are, therefore, limited in size. To compensate for the limited size, these devices often acquire multiple partial impressions of a single finger during enrollment to ensure that at least one of them will successfully match with the image obtained from the user during authentication. Furthermore, in some cases, the user is allowed to enroll multiple fingers, and the impressions pertaining to multiple partial fingers are associated with the same identity (i.e., one user). A user is said to be successfully authenticated if the partial fingerprint obtained during authentication matches any one of the stored templates. This paper investigates the possibility of generating a “MasterPrint,” a synthetic or real partial fingerprint that serendipitously matches one or more of the stored templates for a significant number of users. Our preliminary results on an optical fingerprint data set and a capacitive fingerprint data set indicate that it is indeed possible to locate or generate partial fingerprints that can be used to impersonate a large number of users. In this regard, we expose a potential vulnerability of partial fingerprint-based authentication systems, especially when multiple impressions are enrolled per finger.

Photo response non-uniformity (PRNU) noise-based source attribution is a well-known technique to verify the camera of an image or video. Researchers have proposed various countermeasures to prevent PRNU-based source camera attribution. Forced seam-carving is one such recently proposed counter forensics technique. This technique can disable PRNUbased source camera attribution by forcefully removing seams such that the size of most uncarved image blocks is less than 50 × 50 pixels. In this paper, we show that given multiple seamcarved images from the same camera, source attribution can still be possible even if the size of uncarved blocks in the image is less than the recommended size of 50 × 50 pixels. Theoretical analysis and experiments with multiple cameras demonstrate that the effectiveness of our scheme depends on the number of seams carved from an image and the randomness of the seam positions.

Photo Response Non-Uniformity (PRNU) noise based source attribution is a well known technique to verify the source camera of an anonymous image or video. Researchers have proposed various counter measures to PRNU based source camera attribution. Forced seam-carving is a recently proposed counter forensics measure that was proposed to defeat PRNU based source attribution by disturbing the alignment of PRNU noise patterns. This paper shows that given a multiple number of seam-carved images, source attribution can still be reliably made even if the size of a non-carved image block is less than the recommended size of 50×50.

Increasing cyber-security presents an ongoing challenge to security professionals. Research continuously suggests that online users are a weak link in information security. This research explores the relationship between cyber-security and cultural, personality and demographic variables.

Enterprise networks have been a frequent target of data breaches and sabotage. In a widely used method, attackers establish a foothold in the target network by compromising a single computer or account. They then move laterally between computers to access valuable resources and information located deeper inside the network. To move laterally, attackers often steal valid user credentials. This paper is based on the observation that an attackers' pattern of access characteristics of the stolen credentials in the form ofdeviates from benign patterns and can be used to detect malicious logins. In this paper, we present APT-Hunter1, a visualization tool that helps security analysts to explore login data for discovering patterns and detecting malicious logins. To evaluate the proposed system, a pilot study was conducted over an open dataset of more than one billion logins of an enterprise network, provided by Los Alamos National Lab (LANL). Using APT-Hunter, security analysts (unfamiliar with the dataset) were able to detect 349 of 749 malicious logins related to lateral movements performed by a Red Team during a penetration test conducted at LANL. APT-Hunter is currently deployed in a global financial company and helps security analysts detect account compromises.

The photo response non-uniformity noise associated with an imaging sensor has been shown to be a unique and persistent identifier that can be treated as the sensor's digital fingerprint. The method for attributing an image to a particular camera, however, is not suitable for source identification due to efficiency considerations, which is a one-to-many matching of a single fingerprint against a database of fingerprints. To address this problem, we propose a group-testing approach based on the notion of composite fingerprints (CFs), generated by combining many actual fingerprints together into a single fingerprint. Our technique organizes a database of fingerprints into an unordered binary search tree, wherein each internal node is represented by a fingerprint composited from all the fingerprints at the leaf nodes in the subtree beneath that node. Different search strategies are considered, and the performance is analyzed analytically and verified using numerical simulations as well as experimental results. Our results are presented in comparison with the linear search-based approach that utilizes fingerprint digests for more effective computation. Results obtained under the best achievable accuracy showed that the proposed method yields a lower overall computational cost. It is also shown that by complementary use of the fingerprint dimension reduction and CF-based search tree approaches, it is possible to further improve the search efficiency.

This paper studies continuous authentication for touch interface based mobile devices. A Hidden Markov Model (HMM) based behavioral template training approach is presented, which does not require training data from other subjects other than the owner of the mobile. The stroke patterns of a user are modeled using a continuous left-right HMM. The approach models the horizontal and vertical scrolling patterns of a user since these are the basic and mostly used interactions on a mobile device. The effectiveness of the proposed method is evaluated through extensive experiments using the Toucha-lytics database which comprises of touch data over time. The results show that the performance of the proposed approach is better than the state-of-the-art method.

PIN authentication is widely used thanks to its simplicity and usability, but it is known to be susceptible to shoulder surfing. In this paper, we propose a novel online finger-drawn PIN authentication technique that lets a user draw a PIN on a touch interface with her finger. The system provides some resilience to shoulder surfing without increasing authentication delay and complexity by using both the PIN as well as a behavioral biometric in user verification. Our approach adopts the Dynamic Time Warping (DTW) algorithm to compute dissimilarity scores between PIN samples. We evaluate our system in two shoulder surfing scenarios: 1) PIN attack where the attacker only knows the victim's PIN but has no information about it's drawing characteristic and 2) Imitation attack where an attacker has access to a dynamic drawing sequence of a victim's finger-drawn PIN in the form of multiple observations. Experimental results with a data set of 40 users and 2400 imitating samples from two attacks yield an Equal Error Rate (EER) of 6.7% and 9.9% respectively, indicating the need for further study on this promising authentication mechanism.

This paper presents a child pornographic image detection system that identifies human skin tones in digital images, extracts features to detect explicit images and performs facial image based age classification. The novelty of the technique relies on the use of a robust and very fast skin color filter and a new set of facial features for improved identification of child faces. Tests on a dataset containing explicit images taken under different illuminations and reflecting a diversity of human skin tones, show that explicit images can be differentiated from benign images with around 90% accuracy. Similarly, tests performed on adult and child facial images yielded an accuracy of 80% in detecting child faces. Test conducted on 105 images involving semi-naked children (with no sexual context) revealed that the system has true positive rates of 83% in detecting explicit-like images and 96.5% in detecting child faces.

We present a semi-automatic method to reconstruct shredded documents. The novelty of the method lies in the way it performs pairwise matching of chads. The technique divides chad contours into curves using corner detection and introduces a procedure to assess the match of two curves. The proposed curve matching technique is robust to translation and rotation and can cope with shape deformations due to shredding by allowing overlapping of chads during matching. The alignment of text lines, crossing characters and color information on the chads is also utilized to improve matching performance. Visual interfaces are designed to allow for user input in identifying correctly matching chad pairs and reconstructing the document. The effectiveness of the method is demonstrated by solving the first and second puzzles of the DARPA shredder challenge.

Automatic expert assignment is a common problem encountered in both industry and academia. For example, for conference program chairs and journal editors, in order to collect "good" judgments for a paper, it is necessary for them to assign the paper to the most appropriate reviewers. Choosing appropriate reviewers of course includes a number of considerations such as expertise and authority, but also diversity and avoiding conflicts. In this paper, we explore the expert retrieval problem and implement an automatic paper-reviewer recommendation system that considers aspects of expertise, authority, and diversity. In particular, a graph is first constructed on the possible reviewers and the query paper, incorporating expertise and authority information. Then a Random Walk with Restart (RWR) [1] model is employed on the graph with a sparsity constraint, incorporating diversity information. Extensive experiments on two reviewer recommendation benchmark datasets show that the proposed method obtains performance gains over state-of-the-art reviewer recommendation systems in terms of expertise, authority, diversity, and, most importantly, relevance as judged by human experts.

The availability of sophisticated source attribution techniques raises new concerns about privacy and anonymity of photographers, activists, and human right defenders who need to stay anonymous while spreading their images and videos. Recently, the use of seam-carving, a content-aware resizing method, has been proposed to anonymize the source camera of images against the well-known photoresponse nonuniformity (PRNU)-based source attribution technique. In this paper, we provide an analysis of the seam-carving-based source camera anonymization method by determining the limits of its performance introducing two adversarial models. Our analysis shows that the effectiveness of the deanonymization attacks depend on various factors that include the parameters of the seam-carving method, strength of the PRNU noise pattern of the camera, and an adversary's ability to identify uncarved image blocks in a seam-carved image. Our results show that, for the general case, there should not be many uncarved blocks larger than the size of $50\times 50$ pixels for successful anonymization of the source camera.

This paper studies online signature verification on touch interface-based mobile devices. A simple and effective method for signature verification is developed. An online signature is represented with a discriminative feature vector derived from attributes of several histograms that can be computed in linear time. The resulting signature template is compact and requires constant space. The algorithm was first tested on the well-known MCYT-100 and SUSIG data sets. The results show that the performance of the proposed technique is comparable and often superior to state-of-the-art algorithms despite its simplicity and efficiency. In order to test the proposed method on finger drawn signatures on touch devices, a data set was collected from an uncontrolled environment and over multiple sessions. Experimental results on this data set confirm the effectiveness of the proposed algorithm in mobile settings. The results demonstrate the problem of within-user variation of signatures across multiple sessions and the effectiveness of cross session training strategies to alleviate these problems.

This paper investigates multitouch gestures for user authentication on touch sensitive devices. A canonical set of 22 multitouch gestures was defined using characteristics of hand and finger movement. Then, a multitouch gesture matching algorithm robust to orientation and translation was developed. Two different studies were performed to evaluate the concept. First, a single session experiment was performed in order to explore feasibility of multitouch gestures for user authentication. Testing on the canonical set showed that the system could achieve good performance in terms of distinguishing between gestures performed by different users. In addition, the tests demonstrated a desirable alignment of usability and security as gestures that were more secure from a biometric point of view were rated as more desirable in terms of ease, pleasure, and excitement. Second, a study involving a three-session experiment was performed. Results indicate that biometric information gleaned from a short user-device interaction remains consistent across gaps of several days, though there is noticeable degradation of performance when the authentication is performed over multiple sessions. In addition, the study showed that user-defined gestures yield the highest recognition rate among all other gestures, whereas the use of multiple gestures in a sequence aids in boosting verification accuracy. In terms of memorability, the study showed that it is feasible for a user to recall user-defined gestural passwords and it is observed that the recall rate increases over time. It is also noticed that performing a user-defined gesture over a customized background image does result in higher verification performance. In terms of usability, the study shows that users did not have difficulty in performing multitouch gestures as they all rated each gesture as easy to perform.

As image source attribution techniques have become significantly sophisticated and are now becoming commonplace, there is a growing need for capabilities to anonymize images and videos. Focusing on the photo response non-uniformity noise pattern based sensor fingerprinting technique, this work evaluates the effectiveness of well-established seam carving method to defend against sensor fingerprint matching. We consider ways in which seam-carving based anonymization can be countered and propose enhancements over conventional seam carving method. Our results show that applying geometrical distortion in addition to seam carving will make counter attack very ineffective both in terms of computational complexity and accuracy.

In this paper, we present a novel multi-touch gesture-based authentication technique. We take advantage of the multi-touch surface to combine biometric techniques with gestural input. We defined a comprehensive set of five-finger touch gestures, based upon classifying movement characteristics of the center of the palm and fingertips, and tested them in a user study combining biometric data collection with usability questions. Using pattern recognition techniques, we built a classifier to recognize unique biometric gesture characteristics of an individual. We achieved a 90% accuracy rate with single gestures, and saw significant improvement when multiple gestures were performed in sequence. We found user ratings of a gestures desirable characteristics (ease, pleasure, excitement) correlated with a gestures actual biometric recognition rate - that is to say, user ratings aligned well with gestural security, in contrast to typical text-based passwords. Based on these results, we conclude that multi-touch gestures show great promise as an authentication mechanism.

It is now established that photo-response nonuniformity noise pattern can be reliably used as a fingerprint to identify an image sensor. The large size and random nature of sensor fingerprints, however, make them inconvenient to store. Further, associated fingerprint matching method can be computationally expensive, especially for applications that involve large-scale databases. To address these limitations, we propose to represent sensor fingerprints in binary-quantized form. It is shown through both analytical study and simulations that the reduction in matching accuracy due to quantization is insignificant as compared to conventional approaches. Experiments on actual sensor fingerprint data are conducted to confirm that only a slight increase occurred in the probability of error and to demonstrate the computational efficacy of the approach.

In this work we show that once a single peer-to-peer (P2P) bot is detected in a network, it may be possible to efficiently identify other members of the same botnet in the same network even before they exhibit any overtly malicious behavior. Detection is based on an analysis of connections made by the hosts in the network. It turns out that if bots select their peers randomly and independently (i.e. unstructured topology), any given pair of P2P bots in a network communicate with at least one mutual peer outside the network with a surprisingly high probability. This, along with the low probability of any other host communicating with this mutual peer, allows us to link local nodes within a P2P botnet together. We propose a simple method to identify potential members of an unstructured P2P botnet in a network starting from a known peer. We formulate the problem as a graph problem and mathematically analyze a solution using an iterative algorithm. The proposed scheme is simple and requires only flow records captured at network borders. We analyze the efficacy of the proposed scheme using real botnet data, including data obtained from both observing and crawling the Nugache botnet.

In this work we propose a Voice over IP (VoIP) call tracking scheme based on robust hashing of VoIP flows. In the proposed scheme the audio content of a possibly encrypted VoIP flow is identified by a short binary string, called the robust hash, using variations on the flow's bitrate over time. These robust hashes are then used to detect pairs of parties communicating with each other. In summary, if two parties are communication with each other then they have a pair of VoIP flows having similar robust hashes with high probability. The basic intuition behind the proposed hash function is twofolds: i)The variable bitrate codec employed in most VoIP applications result in a distinctive bitrate variations over time for each VoIP flow depending on the underlying audio content. ii) Encryption typically doesn't change the bitrate of a VoIP flow. Our experiments show that the proposed scheme is able to identify Skype VoIP flows even under various network impairments such as packet delays, jitter and packet drops.

Most work on steganalysis, except a few exceptions, have primarily focused on providing features with high discrimination power without giving due consideration to issues concerning practical deployment of steganalysis methods. In this work, we focus on machine learning aspect of steganalyzer design and utilize a hierarchical ensemble of classifiers based approach to tackle two main issues. Firstly, proposed approach provides a workable and systematic procedure to incorporate several steganalyzers together in a composite steganalyzer to improve detection performance in a scalable and cost-effective manner. Secondly, since the approach can be readily extended to multi-class classification it can also be used to infer the steganographic technique deployed in generation of a stego-object. We provide results to demonstrate the potential of the proposed approach.

Being able to measure the actual information content of biometrics is very important but also a challenging problem. Main difficulty here is not only related to the selected feature representation of the biometric data, but also related to the matching algorithm employed in biometric systems. In this paper, we propose a new measure for measuring biometric information using relative entropy between intra-user and inter-user distance distributions. As an example, we evaluated the proposed measure on a face image dataset.

We present an efficient and robust stepping-stone detection scheme based on succinct packet-timing sketches of network flows. The proposed scheme employs an online algorithm to continuously maintain short sketches of flows from a stream of captured packets at the network boundary. These sketches are then used to identify pairs of network flows with similar packet-timing characteristics, which indicates potential stepping-stones. Succinct flow sketches enable the proposed scheme to compare a given pair of flows in constant time. In addition, flow sketches identify pairs of correlated flows from a given list of flows in sub-quadratic time, thereby allowing a more scalable solution as compared to known schemes. Finally, the proposed scheme is resistant to random delays and chaff, which are often employed by attackers to evade detection. To explore its efficacy, we mathematically analyze the robustness properties of the proposed flow sketch. We also experimentally measure the detection performance of the proposed scheme.

In this paper we propose a novel out-of-core technique for progressive lossless compression and selective decompression of 3D triangle meshes larger than main memory. Most existing compression methods, in order to optimize compression ratios, only allow sequential decompression. We develop an integrated approach that resolves the issue of so-called prefix dependency to support selective decompression, and in addition enables I/O-efficient compression, while maintaining high compression ratios. Our decompression scheme initially provides a global context of the entire mesh at a coarse resolution, and allows the user to select different regions of interest to further decompress/refine to different levels of details, to facilitate out-of-core multiresolution rendering for interactive visual inspection. We present experimental results which show that we achieve fast compression/decompression times and low memory footprints, with compression ratios comparable to current out-of-core single resolution methods.

Copy-move forgery is a specific type of image tampering, where a part of the image is copied and pasted on another part of the same image. In this paper, we propose a new approach for detecting copy-move forgery in digital images, which is considerably more robust to lossy compression, scaling and rotation type of manipulations. Also, to improve the computational complexity in detecting the duplicated image regions, we propose to use the notion of counting bloom filters as an alternative to lexicographic sorting, which is a common component of most of the proposed copy-move forgery detection schemes. Our experimental results show that the proposed features can detect duplicated region in the images very accurately, even when the copied region was undergone severe image manipulations. In addition, it is observed that use of counting bloom filters offers a considerable improvement in time efficiency at the expense of a slight reduction in the robustness.

Cooperative wireless communication protocols are designed with the assumption that users always behave in a socially efficient manner. This assumption may not be valid in commercial wireless networks where users may violate rules of cooperation to reap benefits of cooperation at no cost. Disobeying the rules of cooperation creates a social-dilemma where well-behaved users exhibit uncertainty about intention of other users. Cooperation in social-dilemma is characterized by a non-cooperative Nash equilibrium which indicates the difficulty of maintaining a socially optimal cooperation without establishing a mechanism to detect and mitigates effects of misbehavior. In this paper, we formulate interaction of users in cooperative Amplify-and-Forward as a dynamic game with incomplete information. We show the existence of a perfect Bayesian equilibrium.

In this paper, a novel individual source scanner identification scheme is proposed. The scheme uses traces of dust, dirt, and scratches over scanner platen on scanned images to characterize a source scanner. The efficacy of the proposed scheme is substantiated with experimental analysis. The robustness of the scheme to the JPEG compression is also investigated. Experimental results show that proposed scheme could be used to match a scanned image to its source.

Computational cost of classification is as important as accuracy in on-line classification systems. The computational cost is usually dominated by the cost of computing implicit features of the raw input data. Very few efforts have been made to design classifiers which perform effectively with limited computational power; instead, feature selection is usually employed as a pre-processing step to reduce the cost of running traditional classifiers. We present CoCoST, a novel and effective approach for building classifiers which achieve state-of-the-art classification accuracy, while keeping the expected computational cost of classification low, even without feature selection. CoCost employs a wide range of novel cost-aware decision trees, each of which is tuned to specialize in classifying instances from a subset of the input space, and judiciously consults them depending on the input instance in accordance with a cost-aware meta-classifier. Experimental results on a network flow detection application show that, our approach can achieve better accuracy than classifiers such as SVM and random forests, while achieving 75%-90% reduction in the computational costs.

In this paper, we introduce tamper detection techniques based on artifacts created by color filter array (CFA) processing in most digital cameras. The techniques are based on computing a single feature and a simple threshold based classifier. The efficacy of the approach was tested over thousands of authentic, tampered, and computer generated images. Experimental results demonstrate reasonably low error rates.

The identification of image acquisition source is an important problem in digital image forensics. In this work, we focus on building a classifier to effectively distinguish between digital images taken from digital single lens reflex (DSLR) and compact cameras. Based on the architecture and the imaging features of DSLR and compact cameras, the images taken from different sources may have different statistical properties in both spatial and transform domains. In this work, we utilized wavelet coefficients and pixel noise statistics to model these two different source classes over 20 different digital cameras. The efficacy of the digital source class identifier, introduced in the paper, has been tested over 1000 high quality camera outputs and post-processed images (resized, re-compressed). Experimental analysis shows that the proposed method has good potential to distinguish DSLR and compact source classes.

Existing cooperative diversity protocols are designed with the inherent assumption that users exhibit cooperative behavior all the time. However, in a practical cooperative wireless system users may misbehave in malicious or selfish manner. Thus existing cooperative diversity protocols are inherently vulnerable to misbehaving users as they lack a mechanism to detect the presence of such users. In this paper we examine the physical layer consequences of a malicious user which exhibits cooperative behavior in a stochastic manner. We assume that the malicious user exploits the inherent uncertainty of the wireless channel to hide its malicious behavior. We consider a malicious user which exhibits cooperative and malicious behaviors according to first- order Markov chain. By behaving stochastically the malicious user attempts to mimic the underlying Markov property of Rayleigh fading channels. Based on this model we examine physical layer performance of cooperative Detect-and-Forward (DF). We show that a malicious user incurs significant degradation in cooperative diversity gain. Our results indicate that misbehaving users may pose formidable challenge to practical implementation of cooperative diversity. Hence, it may be difficult to implement practical wireless cooperative networks without a mechanism to ensure cooperation.

Digital single lens reflex cameras suffer from a well-known sensor dust problem due to interchangeable lenses that they deploy. The dust particles that settle in front of the imaging sensor create a persistent pattern in all captured images. In this paper, we propose a novel source camera identification method based on detection and matching of these dust-spot characteristics. Dust spots in the image are detected based on a (Gaussian) intensity loss model and shape properties. To prevent false detections, lens parameter-dependent characteristics of dust spots are also taken into consideration. Experimental results show that the proposed detection scheme can be used in identification of the source digital single lens reflex camera at low false positive rates, even under heavy compression and downsampling.

Relay nodes are a potential threat to networks since they are used in many malicious situations like stepping stone attacks, botnet communication, peer-to-peer streaming etc. Quick and accurate detection of relay nodes in a network can significantly improve security policy enforcement. There has been significant work done and novel solutions proposed for the problem of identifying relay flows active within a node in the network. However, these solutions require quadratic number of comparisons in the number of flows. In this paper, a related problem of identifying relay nodes is investigated where a relay node is defined as a node in the network that has an active relay flow. The problem is formulated as a variance estimation problem and a statistical approach is proposed for the solution. The proposed solution requires linear time and space in the number of flows and therefore can be employed in large scale implementations. It can be used on its own to identify relay nodes or as a first step in a scalable relay flow detection solution that performs known quadratic time analysis techniques for relay flow detection only on nodes that have been detected as relay nodes. Experimental results show that the proposed scheme is able to detect relay nodes even in the presence of intentional inter-packet delays and chaff packets introduced by adversaries in order to defeat timing based detection algorithms.

Since biometric data cannot be easily replaced or revoked, it is important that biometric templates used in biometric applications should be constructed and stored in a secure way, such that attackers would not be able to forge biometric data easily even when the templates are compromised. This is a challenging goal since biometric data are "noisy" by nature, and the matching algorithms are often complex, which make it difficult to apply traditional cryptographic techniques, especially when multiple modalities are considered. In this paper, we consider a "fusion " of a minutiae-based fingerprint authentication scheme and an SVD-based face authentication scheme, and show that by employing a recently proposed cryptographic primitive called "secure sketch ", and a known geometric transformation on minutiae, we can make it easier to combine different modalities, and at the same time make it computationally infeasible to forge an "original" combination of fingerprint and face image that passes the authentication. We evaluate the effectiveness of our scheme using real fingerprints and face images from publicly available sources.

Discrimination of computer generated images from real images is becoming more and more important. In this paper, we propose the use of new features to distinguish computer generated images from real images. The proposed features are based on the differences in the acquisition process of images. More specifically, traces of demosaicking and chromatic aberration are used to differentiate computer generated images from digital camera images. It is observed that the former features perform very well on high quality images, whereas the latter features perform consistently across a wide range of compression values. The experimental results show that proposed features are capable of improving the accuracy of the state-of-the-art techniques.

Designing secure digital watermarking schemes resistant to invertibility attacks (or more generally, ambiguity attacks) has been challenging. In a recent work, Li and Chang (IHW'04) give the first stand-alone provably secure non-invertible spread-spectrum watermarking scheme based on cryptographically secure pseudo-random generators. Despite its provable security, there are certain constraints on the security parameters that require further analysis in practice, where it is more important to analyze the exact security instead of theoretical asymptotic bounds. In this paper, we consider a security notion that is slightly weaker theoretically but still reasonable in practice, and show that with this alternative security notion, the exact requirements on the parameters can be analyzed, and such analysis can be used to guide flexible implementations of similar schemes in practice.

Powerful digital media editing tools make producing good quality forgeries very easy for almost anyone. Therefore, proving the authenticity and integrity of digital media becomes increasingly important. In this work, we propose a simple method to detect image tampering operations that involve sharpness/blurriness adjustment. Our approach is based on the assumption that if a digital image undergoes a copy-paste type of forgery, average sharpness/blurriness value of the forged region is expected to be different as compared to the non-tampered parts of the image. The method of estimating sharpness/blurriness value of an image is based on the regularity properties of wavelet transform coefficients which involves measuring the decay of wavelet transform coefficients across scales. Our preliminary results show that the estimated sharpness/blurriness scores can be used to identify tampered areas of the image.

In this paper, we propose a novel method for robust document image authentication. In the proposed method, characters and symbols on a binary document are first grouped into different classes based on k-means clustering in the feature space. Labels are then assigned to the different classes. An ordered sequence of labels formed from the characters and symbols on the document is then used to compute a digital signature. This is achieved by using a cryptographic hash function and a secret key. The computed signature can be appended at the end of an electronic document file, or printed on a hardcopy document in the form of a bar code. Using this method, we will be able to detect any intentional content alteration to the document, but at the same time tolerate moderate amount of noise introduced by printing, scanning, and photocopying of the document. Experimental results demonstrate the effectiveness of the proposed approach.

In a novel method for identifying the source camera of a digital image is proposed. The method is based on first extracting imaging sensor's pattern noise from many images and later verifying its presence in a given image through a correlative procedure. In this paper, we investigate the performance of this method in a more realistic setting and provide results concerning its detection performance. To improve the applicability of the method as a forensic tool, we propose an enhancement over it by also verifying that class properties of the image in question are in agreement with those of the camera. For this purpose, we identify and compare characteristics due to demosaicing operation. Our results show that the enhanced method offers a significant improvement in the performance.

We develop a model to identify the most likely regions for users to click in order to create graphical passwords in the PassPoints system. A PassPoints password is a sequence of points, chosen by a user in an image that is displayed on the screen. Our model predicts probabilities of likely click points; this enables us to predict the entropy of a click point in a graphical password for a given image. The model allows us to evaluate automatically whether a given image is well suited for the PassPoints system, and to analyze possible dictionary attacks against the system. We compare the predictions provided by our model to results of experiments involving human users. At this stage, our model and the experiments are small and limited; but they show that user choice can be modeled and that expansions of the model and the experiments are a promising direction of research.

Practical applications that employ entropy coding for large alphabets often partition the alphabet set into two or more layers, and encode each symbol by using some suitable prefix coding for each layer. In this paper, we formulate the problem of finding an alphabet partitioning for the design of a two-layer semiadaptive code as an optimization problem, and give a solution based on dynamic programming. However, the complexity of the dynamic programming approach can be quite prohibitive for a long sequence and a very large alphabet size. Hence, we also give a simple greedy heuristic algorithm whose running time is linear in the length of the input sequence, irrespective of the underlying alphabet size. Although our dynamic programming and greedy algorithms do not provide a globally optimal solution for the alphabet partitioning problem, experimental results demonstrate that superior prefix coding schemes for large alphabets can be designed using our new approach

This paper focuses on a problem that is common to most watermarking-based ownership dispute resolutions and ownership assertion systems. Such systems are vulnerable to a simple but effective class of attacks that exploit the high false-positive rate of the watermarking techniques to cast doubt on the reliability of a resulting decision. To mitigate this vulnerability, we propose embedding multiple watermarks, as opposed to embedding a single watermark, and detecting a randomly selected subset of them while constraining the embedding distortion. The crux of the scheme lies in both watermark generation, which deploys a family of one-way functions and selective detection, which injects uncertainty into the detection process. The potential of this approach in reducing the false-positive probability is analyzed under various operating conditions and compared to single watermark embedding. The multiple watermark embedding and selective detection technique is incorporated analytically into the additive watermarking technique and results obtained through numerical solutions are presented to illustrate its effectiveness.

Secure storage of biometric templates has become an increasingly important issue in biometric authentication systems. We study how secure sketch, a recently proposed error-tolerant cryptographic primitive, can be applied to protect the templates. We identify several practical issues that are not addressed in the existing theoretical framework, and show the subtleties in evaluating the security of practical systems. We propose a general framework to design and analyze a secure sketch for biometric templates, and give a concrete construction for face biometrics as an example. We show that theoretical bounds have their limitations in practical schemes, and the exact security of the system often needs more careful investigations. We further discuss how to use secure sketch in the design of multifactor authentication systems that allow easy revocation of user credentials.

We describe a digital image forensics technique to distinguish images captured by a digital camera from computer generated images. Our approach is based on the fact that image acquisition in a digital camera is fundamentally different from the generative algorithms deployed by computer generated imagery. This difference is captured in terms of the properties of the residual image (pattern noise in case of digital camera images) extracted by a wavelet based denoising filter. In (Jan Lukas, et al., 2005), it is established that each digital camera has a unique pattern noise associated with itself. In addition, our results indicate that the two type of residuals obtained from different digital camera images and computer generated images exhibit some common characteristics that is not present in the other type of images. This can be attributed to fundamental differences in the image generation processes that yield the two types of images. Our results are based on images generated by the Maya and 3D Studio Max software, and various digital camera images.

The primary goal of image steganography techniques has been to maximize embedding rate while minimizing the detectability of the resulting stego images against steganalysis techniques. However, one particular advantage of steganography, as opposed to other information hiding techniques, is that the embedder has the freedom to choose a cover image that result in the least detectable stego image. This resource has largely remained unexploited in the proposed embedding techniques. In this paper, we study the problem of cover selection by investigating three scenarios in which the embedder has either no knowledge, partial knowledge, or full knowledge of the steganalysis technique. For example, we illustrate through experiments how simple statistical measures could help embedder minimize detectability, at times by 65%, in the partial knowledge case.

In this paper we investigate the problem of lossless geometry compression of irregular-grid volume data represented as a tetrahedral mesh. We propose a novel lossless compression technique that effectively predicts, models, and encodes geometry data for both steady-state (i.e., with only a single time step) and time-varying datasets. Our geometry coder is truly lossless and also does not need any connectivity information. Moreover, it can be easily integrated with a class of the best existing connectivity compression techniques for tetrahedral meshes with a small amount of overhead information. We present experimental results which show that our technique achieves superior compression ratios, with reasonable encoding times and fast (linear) decoding times.

This paper generalizes Blonder's graphical passwords to arbitrary images and solves a robustness problem that this generalization entails. The password consists of user-chosen click points in a displayed image. In order to store passwords in cryptographically hashed form, we need to prevent small uncertainties in the click points from having any effect on the password. We achieve this by introducing a robust discretization, based on multigrid discretization.

The problem of restoring deleted files from a scattered set of fragments arises often in digital forensics. File fragmentation is a regular occurrence in hard disks, memory cards, and other storage media. As a result, a forensic analyst examining a disk may encounter many fragments of deleted digital files, but is unable to determine the proper sequence of fragments to rebuild the files. In this paper, we investigate the specific case where digital images are heavily fragmented and there is no file table information by which a forensic analyst can ascertain the correct fragment order to reconstruct each image. The image reassembly problem is formulated as a k-vertex disjoint graph problem and reassembly is then done by finding an optimal ordering of fragments. We provide techniques for comparing fragments and describe several algorithms for image reconstruction based on greedy heuristics. Finally, we provide experimental results showing that images can be reconstructed with high accuracy even when there are thousands of fragments and multiple images involved.

Identification and verification of a video clip via its fingerprint find applications in video browsing, database search and security. For this purpose, the video sequence must be collapsed into a short fingerprint using a robust hash function based on signal processing operations. We propose two robust hash algorithms for video based both on the discrete cosine transform (DCT), one on the classical basis set and the other on a novel randomized basis set (RBT). The robustness and randomness properties of the proposed hash functions are investigated in detail. It is found that these hash functions are resistant to signal processing and transmission impairments, and therefore can be instrumental in building database search, broadcast monitoring and watermarking applications for video. The DCT hash is more robust, but lacks security aspect, as it is easy to find different video clips with the same hash value. The RBT based hash, being secret key based, does not allow this and is more secure at the cost of a slight loss in the receiver operating curves

In this paper we propose a novel geometry compression technique for 3D triangle meshes. We focus on a commonly used technique for predicting vertex positions via a flipping operation using the parallelogram rule. We show that the efficiency of the flipping operation is dependent on the order in which triangles are traversed and vertices are predicted accordingly. We formulate the problem of optimally (traversing triangles and) predicting the vertices via flippings as a combinatorial optimization problem of constructing a constrained minimum spanning tree. We give heuristic solutions for this problem and show that we can achieve prediction efficiency within 17.4% on average as compared to the unconstrained minimum spanning tree which is an unachievable lower bound. We also show significant improvements over previous techniques in the literature that strive to find good traversals that also attempt to minimize prediction errors obtained by a sequence of flipping operations, albeit using a different approach.

Watermarking technologies have been envisioned as a potential means for establishing ownership on digital media objects. However, achievable robustness and false-positive rates of the state-of-the-art watermarking techniques raise doubts about applicability of watermarking to ownership problem. With this perspective, we address the security weaknesses common to most watermarking techniques and assess the role of watermarking in construction of ownership assertion systems. We identify the requirements of a watermarking based ownership assertion system. Also, we provide a basic functional outline of a practical version of such a system and identify its potential vulnerabilities. To mitigate these vulnerabilities, we aim at reducing the false positive rate of the watermark detection scheme. For this purpose, we propose embedding multiple watermarks as opposed to single watermark embedding while constraining the embedding distortion. The crux of the proposed method lies in watermark generation which deploys a family of one-way functions. We incorporate the multiple watermark embedding idea with the additive watermarking technique [1] and present results to illustrate the potential of this approach in reducing the false-positive rate of the watermark detection scheme.

We present a novel steganalysis technique for halftone images without knowledge of the original cover image. We first convert halftone images into grayscale-like images by low-pass filtering. The low-pass-filtered image is then decomposed using quadrature mirror filters, and a set of subband coefficients are generated at different scales and orientations. Next, a set of statistical features are computed from the subband coefficients and their predicated errors. Using Fisher linear discriminant analysis, a statistical classifier is designed to detect marked images. Experimental results demonstrate the effectiveness and accuracy of the proposed technique.

In this paper, we propose a novel approach to image steganography in which embedding is done without making explicit modifications to the image; that is, the embedding distortion introduced to the cover image is both perceptually and statistically ensured to be less detectable. If an image is divided into blocks and each block is hashed, then the hash values could represent the embedded message content. A set of replacement blocks can be obtained by consecutively capturing images of the same scene (or resampling the incident light). Since image content remains the same and noise is sampled, the set of replacements are statistically compatible while still providing unique hash values. With such an approach, the embedder can choose the block with hash value corresponding to the message content without violating any of the natural image statistics. When applied to JPEG images, experiments show that this technique can achieve embedding rates of 0.063 bits per DCT coefficient or 0.157 bits per nonzero DCT coefficient while still remaining undetectable by Farid's universal steganalysis tool.

In this work, we focus our interest on blind source camera identification problem by extending our results in the direction of M. Kharrazi et al. (2004). The interpolation in the color surface of an image due to the use of a color filter array (CFA) forms the basis of the paper. We propose to identify the source camera of an image based on traces of the proprietary interpolation algorithm deployed by a digital camera. For this purpose, a set of image characteristics are defined and then used in conjunction with a support vector machine based multi-class classifier to determine the originating digital camera. We also provide initial results on identifying source among two and three digital cameras.

In this paper, a steganalysis technique using compression bit rate as a distinguishing statistic is presented to detect secret messages embedded in document images that are degraded in quality by printing, photocopying, and/or scanning processes. We consider embedding techniques that flip pixels in binary document images that contain characters and symbols. Noise introduced by printing, photocopying, and/or scanning can be modeled by a local optical distortion process. Steganographic embedding is modeled as an additive noise process and we use compression bit rate as a distinguishing statistic to discriminate between stego images and unmarked images. Experimental results showed that the proposed technique can detect stego images with reasonably good accuracy, given the inherent difficulty of the problem

Graphical passwords are an alternative to alphanumeric passwords in which users click on images to authenticate themselves rather than type alphanumeric strings. We have developed one such system, called PassPoints, and evaluated it with human users. The results of the evaluation were promising with respect to rmemorability of the graphical password. In this study we expand our human factors testing by studying two issues: the effect of tolerance, or margin of error, in clicking on the password points and the effect of the image used in the password system. In our tolerance study, results show that accurate memory for the password is strongly reduced when using a small tolerance (10 x 10 pixels) around the user's password points. This may occur because users fail to encode the password points in memory in the precise manner that is necessary to remember the password over a lapse of time. In our image study we compared user performance on four everyday images. The results indicate that there were few significant differences in performance of the images. This preliminary result suggests that many images may support memorability in graphical password systems.

We introduce hashed random preloaded subsets (HARPS), a highly scalable key predistribution (KPD) scheme employing only symmetric cryptographic primitives. HARPS is ideally suited for resource constrained nodes that need to operate for extended periods without active involvement of a trusted authority (TA), as is usually the case for nodes forming ad hoc networks (AHNs). HARPS, a probabilistic KPD scheme, is a generalization of two other probabilistic KPDs. The first, random preloaded subsets (RPSs), is based on random intersection of keys preloaded in nodes. The second, proposed by Leighton and Micali (LM) is a scheme employing repeated applications of a cryptographic hash function. We investigate many desired properties of HARPS like scalability, computational and storage efficiency, flexibility in deployment modes, renewability, ease of extension to multicast scenarios, ability to cater for broadcast authentication, broadcast encryption, etc., to support its candidacy as an enabler for ad hoc network security. We analyze and compare the performance of the three schemes and show that HARPS has significant advantages over other KPDs, and in particular, over RPS and LM.

We present a compression technique that provides progressive transmission as well as lossless and near-lossless compression in a single framework. The proposed technique produces a bit stream that results in a progressive, and ultimately lossless, reconstruction of an image similar to what one can obtain with a reversible wavelet codec. In addition, the proposed scheme provides near-lossless reconstruction with respect to a given bound, after decoding of each layer of the successively refinable bit stream. We formulate the image data-compression problem as one of successively refining the probability density function (pdf) estimate of each pixel. Within this framework, restricting the region of support of the estimated pdf to a fixed size interval then results in near-lossless reconstruction. We address the context-selection problem, as well as pdf-estimation methods based on context data at any pass. Experimental results for both lossless and near-lossless cases indicate that the proposed compression scheme, that innovatively combines lossless, near-lossless, and progressive coding attributes, gives competitive performance in comparison with state-of-the-art compression schemes.

The new-found ubiquity of embedded processors in consumer and industrial applications brings with it an intensified focus on security, as a strong level of trust in the system software is crucial to their widespread deployment. The growing area of software protection attempts to address the key steps used by hackers in attacking a software system. In this paper, we introduce a unique approach to embedded software protection that utilizes a hardware/software codesign methodology. Results demonstrate that this framework can be the successful basis for the development of embedded applications that meet a wide range of security and performance requirements.

One of the growing problems faced by network administrators is the abuse of computing resources by authorized and unauthorized personnel. The nature of abuse may vary from using unauthorized applications to serving unauthorized content. Proliferation of peer-to-peer networks and wide use of tunnels makes it difficult to detect such abuses and easy to circumvent security policies. This paper presents the design and implementation of a system, called Nabs, that characterizes content types of network flows based solely on the payload which can then be used to identify abuses of computing resources. The proposed method does not depend on packet headers or other simple packet characteristics hence is more robust to circumvention.

Payload attribution is an important problem often encountered in network forensics. Given an excerpt of a payload, finding its source and destination is useful for many security applications such as identifying sources and victims of a worm or virus. Although IP traceback techniques have been proposed in the literature, these techniques cannot help when we do not have the entire packet or when we only have an excerpt of the payload.

Any key pre-distribution (KPD) scheme is inherently a trade-off between complexity and security. By sacrificing some security (KPD schemes need some assurance of the ability to limit sizes of attacker coalitions), KPD schemes gain many advantages. We argue that random KPD schemes, in general, perform an "advantageous" trade-off which renders them more suitable for practical large scale deployments of resource constrained nodes. We introduce a novel random KPD scheme, hashed random preloaded subsets (HARPS), which turns out to be a generalization of two other random KPD schemes, random preloaded subsets (RPS) and a scheme proposed by T. Leighton and S. Micali (LM). All three schemes have probabilistic measures for the "merit" of the system. We analyze and compare the performance of the three schemes. We show that HARPS has significant advantages over other KPD schemes, and in particular, over RPS and LM.

This paper introduces the measure of information as a new criterion for the performance analysis of routing algorithms in wireless sensor networks. We argue that since the objective of a sensor network is to estimate a two dimensional random field, a routing algorithm must maximize information flow about the underlying field over the lifetime of the sensor network. We develop two novel algorithms, MIR (maximum information routing) and CMIR (conditional maximum information routing) designed to maximize information flow, and present a comparison of the algorithms to a previously proposed algorithm-MREP (maximum residual energy path) through simulations. We show that the proposed algorithms give significant improvement in terms of information flow, when compared to MREP.

We propose a practical steganalysis method to detect hidden information and estimate embedding rate. By modelling steganographic embedding as an additive noise process, we exploit the fact that the mean and variance of stegosignal is a function of embedding rate. This can be used to estimate the embedding rate without knowledge of the original cover object. We present simulation results to demonstrate that the proposed method can estimate embedding rate with reasonable accuracy. The technique we propose is general and can be used for steganalysis of images, video, and audio etc.

An existing technique for robust streaming of 3D graphics contents over lossy networks is multi-resolution coding of 3D geometry. An advantage of this approach is that it uses refinement layers and therefore multiple clients with different bandwidths can be served by a single unified code stream. However, there is a dependency between refinement layers, called prefix condition. Decoding of a given layer requires the knowledge of all the previous layers. A problem in the base layer reception interrupts the streaming all together and voids the remaining layers even though they are received perfectly. To overcome this drawback, this paper proposes an alternative approach to multi-resolution geometry coding, called multiple-description coding of 3D geometry. Instead of organizing code stream into embedded layers, MDC generates several separate descriptions of a geometric object, called co-descriptions. Each co-description of MDC can be independently decoded without any knowledge of other co-descriptions. Each extra successfully received co-description improves the fidelity of reconstructed geometry regardless of what has been received so far or in what order.

We propose a quantitative steganalysis method to detect hidden information embedded by flipping pixels along boundaries in binary images. We model steganographic embedding as an additive noise process and use compression rate as a distinguishing statistic that aids in discriminating between stego-images and cover-images. We specifically use the JBIG 2 binary image compression algorithm to derive a quantitative relation between compression rate and embedding rate. Based on this relationship, a practical steganalysis technique is proposed by examining the change of compression rate as embedding rate increases. Experiments conducted show that the proposed technique can reliably detect a steganographic embedding process that flips boundary pixels. Furthermore, it can estimate embedding rate with reasonable accuracy.

An interesting problem in digital forensics is that given a digital image, would it be possible to identify the camera model which was used to obtain the image. In this paper we look at a simplified version of this problem by trying to distinguish between images captured by a limited number of camera models. We propose a number of features which could be used by a classifier to identify the source camera of an image in a blind manner. We also provide experimental results and show reasonable accuracy in distinguishing images from the two and five different camera models using the proposed features.

In this paper, we present a novel technique for the steganalysis of digital documents, when the secret message is embedded along the boundaries of text characters or other symbols in the documents. The proposed technique uses an auto-regressive model to detect marked documents, as well as to estimate the relative length of the embedded messages. Experimental results demonstrate the effectiveness and accuracy of the proposed technique

We propose a system for digital rights management (DRM) which facilitates large scale deployments of heterogeneous devices, manufactured by different vendors, to interact and authenticate with each other securely in order to ensure fairness of transactions. The proposed system allows for both pluggable (and transferable) security modules for end-user authentication and built-in (non-transferable) security modules for mutual authentication of compliant devices (manufactured by different vendors). For the underlying security mechanism we propose the use of a recently proposed random key predistribution scheme, HARPS (Hashed Random Preloaded Subsets).

Reassembly of fragmented objects from a collection of randomly mixed fragments is a common problem in classical forensics. We address the digital forensic equivalent, i.e., reassembly of document fragments, using statistical modelling tools applied in data compression. We propose a general process model for automatically analyzing a collection fragments to reconstruct the original document by placing the fragments in proper order. Probabilities are assigned to the likelihood that two given fragments are adjacent in the original using context modelling techniques in data compression. The problem of finding the optimal ordering is shown to be equivalent to finding a maximum weight Hamiltonian path in a complete graph. Heuristics are designed and explored and implementation results provided which demonstrate the validity of the proposed technique.

Summary form only given. The error susceptibility in the compressed bit stream is considered as a key limitation of adaptive lossless compression systems. The inherent design of these systems often requires that they discard all data subsequent to the error. This is especially problematic in the Burrows-Wheeler blocksorting transform (BWT), with 1MB suffix-sorted blocks. Error-correcting codes, such as Reed-Solomon codes, can be used but their design allows for a maximum pre-fixed error rate. If the channel errors exceed the maximum pre-fixed error rate, the whole block is lost. An error resilient version of the BWT was presented that has error-free output in low channel noise. It gracefully degrades output quality, as errors increase by scattering output errors, and avoids significant error propagation typical with adaptive lossless compression systems. These techniques give interesting new insights on the increasingly popular compression algorithm.

Practical applications that employ entropy coding for large alphabets often partition the alphabet set into two or more layers. Each symbol was encoded using suitable prefix coding for each layer. The problem of optimal alphabet partitioning was formulated for the design of a two layer semi-adaptive code and the given solution was based on dynamic programming. However, the complexity of the dynamic programming approach can be quite prohibitive for a long sequence and very large alphabet size. Hence, a simple greedy heuristic algorithm whose running time is linear in the number of symbols being encoded was given, irrespective of the underlying alphabet size. The given experimental results demonstrated the fact that superior prefix coding schemes for large alphabets can be designed using this approach as opposed to the typically ad-hoc partitioning approach applied in the literature.

Protection of digital data from unauthorized access is of paramount importance. In the past several years, much research has concentrated on protecting data from the standpoint of confidentiality, integrity and availability. Software is a form of data with unique properties and its protection poses unique challenges. First, software can be reverse engineered, which may result in stolen intellectual property. Second, software can be altered with the intent of performing operations this software must not be allowed to perform.With commercial software increasingly distributed in forms from which source code can be easily extracted, such as Java bytecodes, reverse engineering has become easier than ever. Obfuscation techniques have been proposed to impede illegal reverse engineers. Obfuscations are program transformations that preserve the program functionality while obscuring the code, thereby protecting the program against reverse engineering. Unfortunately, the existing obfuscation techniques are limited to obscuring variable names, transformations of local control flow, and obscuring expressions using variables of primitive types. In this paper, we propose obfuscations of design of object-oriented programs.We describe three techniques for obfuscation of program design. The class coalescing obfuscation replaces several classes with a single class. The class splitting obfuscation replaces a single class with multiple classes, each responsible for a part of the functionality of the original class. The type hiding obfuscation uses the mechanism of interfaces in Java to obscure the types of objects manipulated by the program. We show the results of our initial experiments with a prototype implementation of these techniques. In particular, we shown that the runtime overheads of these obfuscations tend to be small.

We address the problem of reassembly of images from a collection of their fragments. The image reassembly problem is formulated as a combinatorial optimization problem and image assembly is then done by finding an optimal ordering of fragments. We present implementation results showing that images can be reconstructed with high accuracy even when there are thousands of fragments and multiple images involved.

Image transcoding proxies are used to improve Web browsing over low bandwidth networks by adapting content-rich Web images to bandwidth-constrained clients. Such transcoding proxies dynamically analyze, manipulate and transcode images (e.g. quality reduction, down sampling) on the fly enabling significant reductions in download times over low bandwidth links. However, transcoding proxies have scalability problems if the objective policy that decides whether to transcode an image does not take the client load (e.g. number of concurrent clients) into consideration. We show that seemingly intuitive policies that make decisions solely based on whether transcoding yields savings in transmission time tail to scale. Under high load, the average latency perceived by a client can be improved by a factor of about two by taking overall client load into consideration and properly scheduling transcoding operations on a single CPU. We show that an earliest deadline first (EDF) based scheduling policy further improves transcoding performance.

In this paper we address the problem of reassembly of images from a collection of their fragments. The image reassembly problem is formulated as a combinatorial optimization problem and image assembly is then done by finding an optimal ordering of fragments. We present implementation results showing that images can be reconstructed with high accuracy even when there are thousands of fragments and multiple images involved.

We present techniques for steganalysis of images that have been potentially subjected to steganographic algorithms, both within the passive warden and active warden frameworks. Our hypothesis is that steganographic schemes leave statistical evidence that can be exploited for detection with the aid of image quality features and multivariate regression analysis. To this effect image quality metrics have been identified based on the analysis of variance (ANOVA) technique as feature sets to distinguish between cover-images and stego-images. The classifier between cover and stego-images is built using multivariate regression on the selected quality metrics and is trained based on an estimate of the original image. Simulation results with the chosen feature set and well-known watermarking and steganographic techniques indicate that our approach is able with reasonable accuracy to distinguish between cover and stego images.

We present a novel technique for steganalysis of images that have been subjected to least significant bit (LSB) type steganographic algorithms. The seventh and eighth bit planes in an image are used for the computation of several binary similarity measures. The basic idea is that the correlation between the bit planes as well the binary texture characteristics within the bit planes differs between a stego-image and a cover-image. These telltale marks can be used to construct a steganalyzer, that is, a multivariate regression scheme to detect the presence of a steganographic message in an image.

We present an LSB steganalysis technique that can detect the existence of hidden messages that are randomly embedded in the least significant bits of natural continuous-tone images. The technique is inspired by the recent work of J. Fridrich et al. (see Proc. ACM Workshop on Multimedia and Security, p.27-30, 2001) and just like their work, it can also precisely measure the length of the embedded message, even when the hidden message is very short relative to the image size. The key to our success is the formation of some subsets of pixels whose cardinalities change with LSB embedding, and such changes can be precisely quantified under the assumption that the embedded bits are randomly scattered. Interestingly, our study on steganalysis of LSB embedding sheds light on the work of Fridrich et al. on the detection of LSB embedding, and offers an analytical proof of an observation made by them.

Delta compression techniques are commonly used to succinctly represent an updated version of a file with respect to an earlier one. We study the use of delta compression in a somewhat different scenario, where we wish to compress a large collection of (more or less) related files by performing a sequence of pairwise delta compressions. The problem of finding an optimal delta encoding for a collection of files by taking pairwise deltas can be reduced to the problem of computing a branching of maximum weight in a weighted directed graph, but this solution is inefficient and thus does not scale to larger file collections. This motivates us to propose a framework for cluster-based delta compression that uses text clustering techniques to prune the graph of possible pairwise delta encodings. To demonstrate the efficacy of our approach, we present experimental results on collections of Web pages. Our experiments show that cluster-based delta compression of collections provides significant improvements in compression ratio as compared to individually compressing each file or using tar+gzip, at a moderate cost in efficiency.

A novel image compression technique is presented that incorporates progressive transmission and near-lossless compression in a single framework. Experimental performance of the proposed coder proves to be competitive with the state-of-the-art compression schemes.

We investigate the image authentication system SARI, proposed by Lin and Chang (see ibid., vol.11, p.153-68, Feb. 2001), that distinguishes JPEG compression from malicious manipulations. In particular, we took at the image digest component of this system. We show that if multiple images have been authenticated with the same secret key and the digests of these images are known to an attacker, Oscar, then he can cause arbitrary images to be authenticated with this same but unknown key. We show that the number of such images needed by Oscar to launch a successful attack is quite small, making the attack very practical. We then suggest possible solutions to enhance the security of this authentication system.

There have been many techniques for hiding messages in images in such a manner that the alterations made to the image are perceptually indiscernible. However, the question whether they result in images that are statistically indistinguishable from untampered images has not been adequately explored. We look at some specific image based steganography techniques and show that an observer can indeed distinguish between images carrying a hidden message and images which do not carry a message. We derive a closed form expression of the probability of detection and false alarm in terms of the number of bits that are hidden. This leads us to the notion of steganographic capacity, that is, how many bits can we hide in a message without causing statistically significant modifications? Our results are able to provide an upper bound on the this capacity. Our ongoing work relates to adaptive steganographic techniques that take explicit steps to foil the detection mechanisms. In this case we hope to show that the number of bits that can be embedded increases significantly.

Mixed raster content (MRC) is an ITU document compression standard (T.44) specifying both a model for multilayer representation of a compound document, and a set of allowable standardized coders for the individual layers. The model requires decomposition of a document into two image layers and a binary mask layer, but the standard does not recommend any procedure for this task. For best compression results, the decomposition method should be optimized for the layer encoders. In this paper, a high performance MRC compound document codec is presented, where the layer decomposition scheme is matched to the JPEG encoder with arithmetic coding for the foreground and background image layers. JBIG is used to code the mask layer. Integrated noise removal routines enable handling of scanned documents along with electronic ones. Resolution scalable decoding features are also implemented. The page segmenter yields a segmentation mask, which serves to separate text and other features.

We investigate the image authentication system, SARI, proposed by C.Y. Lin and S.F. Chang (see SPIE Storage and Retrieval of Image/Video Databases, 1998), that distinguishes JPEG compression from malicious manipulations. In particular, we look at the image digest component of this system. We show that if multiple images have been authenticated with the same secret key and the digests of these images are known to an attacker, Oscar, then he can cause arbitrary images to be authenticated with this same but unknown key. We show that the number of such images needed by Oscar to launch a successful attack is quite small, making the attack very practical. We then suggest possible solutions to enhance the security of this authentication system.

We present techniques for steganalysis of images that have been potentially subjected to a watermarking algorithm. We show that watermarking schemes leave statistical evidence or structure that can be exploited for detection with the aid of proper selection of image features and multivariate regression analysis. We use some image quality metrics as the feature set to distinguish between watermarked and unwatermarked images and furthermore distinguish between different watermarking techniques. To identify specific quality measures that provide the best discriminative power, we use analysis of variance (ANOVA) techniques. Multivariate regression analysis is then used on the selected quality metrics to build an optimal classifier using a set of test images and their blurred versions. Simulation results with a specific feature set and some well-known and publicly available watermarking techniques indicate that our approach is able to accurately distinguish with high accuracy between images marked by different watermarking techniques.

Digital watermarks have previously been proposed for the purposes of copy protection and copy deterrence for multimedia content. In copy deterrence, a content owner (seller) inserts a unique watermark into a copy of the content before it is sold to a buyer. If the buyer sells unauthorized copies of the watermarked content, then these copies can be traced to the unlawful reseller (original buyer) using a watermark detection algorithm. One problem with such an approach is that the original buyer whose watermark has been found on unauthorized copies can claim that the unauthorized copy was created or caused (for example, by a security breach) by the original seller. In this paper, we propose an interactive buyer-seller protocol for invisible watermarking in which the seller does not get to know the exact watermarked copy that the buyer receives. Hence the seller cannot create copies of the original content containing the buyer's watermark. In cases where the seller finds an unauthorized copy, the seller can identify the buyer from a watermark in the unauthorized copy and furthermore the seller can prove this fact to a third party using a dispute resolution protocol. This prevents the buyer from claiming that an unauthorized copy may have originated from the seller.

We describe a watermarking scheme for ownership verification and authentication. Depending on the desire of the user, the watermark can be either visible or invisible. The scheme can detect any modification made to the image and indicate the specific locations that have been modified. If the correct key is specified in the watermark extraction procedure, then an output image is returned showing a proper watermark, indicating the image is authentic and has not been changed since the insertion of the watermark. Any modification would be reflected in a corresponding error in the watermark. If the key is incorrect, or if the image was not watermarked, or if the watermarked image is cropped, the watermark extraction algorithm will return an image that resembles random noise. Since it requires a user key during both the insertion and the extraction procedures, it is not possible for an unauthorized user to insert a new watermark or alter the existing watermark so that the resulting image will pass the test. We present secret key and public key versions of the technique.

Though most lossless image coding techniques use a raster scan to order the pixels for context-based predictive coding, other scans, such as the Hilbert or Peano scan, have been proposed as alternatives with potentially better performance. However, a general understanding of the merits of different scans has been lacking. In previous work, the authors had presented a framework in which the effect of pixel scan order on lossless compression can be quantitatively analyzed, so that comparisons of different scans can be made. Assuming a quantized-Gaussian and isotropic image model with contexts consisting of previously scanned adjacent pixels in a distance constrained neighborhood, it was found that the raster scan is better than the Hilbert scan. In this paper we further develop our arguments and show that for a large class of scans, which we call axial symmetric scans, the raster scan is indeed optimal. We would like to note that many common scans including the Hilbert scan fall under the class of axial symmetric scans.

In this paper, we describe a class of attacks on certain block-based oblivious watermarking schemes. We show that oblivious watermarking techniques that embed information into a host image in a block-wise independent fashion are vulnerable to a counterfeiting attack. Specifically, given a watermarked image, one can forge the watermark it contains into another image without knowing the secret key used for watermark insertion and in some cases even without explicitly knowing the watermark. We demonstrate successful implementations of this attack on a few watermarking techniques that have been proposed in the literature. We also describe a possible solution to this problem of block-wise independence that makes our attack computationally intractable.

Though most image coding techniques use a raster scan to order pixels prior to coding, Hilbert and other scans have been proposed as having better performance due to their superior locality preserving properties. However, a general understanding of the merits of various scans has been lacking. This paper develops an approach for quantitatively analyzing the effect of pixel scan order for context-based, predictive lossless image compression and uses it to compare raster, Hilbert, random and hierarchical scans. Specifically, for a quantized-Gaussian image model and a given scan order, it shows how the encoding rate can be estimated from the frequencies with which various pixel configurations are available as previously scanned contexts, and from the corresponding conditional differential entropies. Formulas are derived for such context frequencies and entropies. Assuming an isotropic image model and contexts consisting of previously scanned adjacent pixels, it is found that the raster scan is better than the Hilbert scan which is often used in compression applications due to its locality preserving properties. The hierarchical scan is better still, though it is based on nonadjacent contexts. The random scan is the worst of the four considered. Extensions and implications of the results to lossy coding are also discussed.

This paper proposes an interband version of CALIC (context-based, adaptive, lossless image codec) which represents one of the best performing, practical and general purpose lossless image coding techniques known today. Interband coding techniques are needed for effective compression of multispectral images like color images and remotely sensed images. It is demonstrated that CALIC's techniques of context modeling of DPCM errors lend themselves easily to modeling of higher-order interband correlations that cannot be exploited by simple interband linear predictors alone. The proposed interband CALIC exploits both interband and intraband statistical redundancies, and obtains significant compression gains over its intrahand counterpart. On some types of multispectral images, interband CALIC can lead to a reduction in bit rate of more than 20% as compared to intraband CALIC. Interband CALIC only incurs a modest increase in computational cost as compared to intraband CALIC.

In this paper, we study compression techniques for electroencephalograph (EEG) signals. A variety of lossless compression techniques, including compress, gzip, bzip, shorten, and several predictive coding methods, are investigated and compared. The methods range from simple dictionary-based approaches to more sophisticated context modeling techniques. It is seen that compression ratios obtained by lossless compression are limited even with sophisticated context-based bias cancellation and activity-based conditional coding. Though lossy compression can yield significantly higher compression ratios while potentially preserving diagnostic accuracy, it is not usually employed due to legal concerns. Hence, we investigate a near-lossless compression technique that gives quantitative bounds on the errors introduced during compression. It is observed that such a technique gives significantly higher compression ratios (up to 3-bit/sample saving with less than 1% error). Compression results are reported for EEG's recorded under various clinical conditions.

In this paper, we present an interband version of CALIC (context-based adaptive lossless image codec), a lossless image coding technique. It is demonstrated that CALIC's techniques of context-based modeling of images lend themselves easily to modeling of image sequences. The generalized interframe CALIC can exploit both interframe and intraframe statistical redundancies, and obtain significant compression gains over intraframe CALIC. The advantage of interframe CALIC is demonstrated by experimental results on different types of multispectral images.

In this paper we apply some of the Golomb-Rice coding techniques that emerged in JPEG-LS, a new standard for lossless image compression, and apply them towards coding DCT coefficients in the lossy JPEG baseline algorithm. We show that this results in significant improvements in performance with limited impact on computational complexity. In fact, one significant reduction in complexity provided by the proposed techniques is the complete elimination of the Huffman tables in JPEG baseline which can be a bottleneck in hardware implementations. We give simulation results, comparing the performance of the proposed technique to JPEG baseline, JPEG baseline with optimal Huffman coding (two pass) and JPEG arithmetic.

The authors have previously analyzed the performance of context-based lossless image coding techniques in conjunction with the Hilbert and raster scans. The analysis revealed that, under certain reasonable assumptions, the raster scan is indeed better than the Hilbert scan, thereby dispelling the popular notion that using a Hilbert scan would always lead to improved performance. In this paper they apply similar techniques to analyze a random scan as well as a progressive scan (closely based on the commonly used HINT scan). They demonstrate that the progressive scan outperforms the raster scan, but that the expected performance of a random scan is inferior to the other three systematic scans considered so far.

Digital watermarks have previously been proposed for the purpose of copy protection and copy deterrence for multimedia content. Copy deterrence using digital watermarks is achieved by inserting a unique watermark into each copy of the watermark content sold which could be used to trace unauthorized copies to the erring buyer. One problem with such an approach is the fact that the buyer whose watermark has been found on unauthorized copies can claim that the unauthorized copy was created or caused (for example, by a security breach) by the seller. In this paper we propose an interactive buyer-seller protocol for invisible watermarking in which the seller does not get to know the exact watermarked copy that the buyer receives. Hence the seller cannot create copies of the original content containing the buyers watermark. However, in case the seller finds an unauthorized copy, he/she can identify the buyer from whom this unauthorized copy has originated and furthermore can also prove this fact to a third party by means of a dispute resolution protocol. Hence, the buyer cannot claim that an unauthorized copy may have originated from the seller.

Digital watermarks have been proposed as a means for copyright protection of multimedia data. We address the capability of invisible watermarking schemes to resolve copyright ownership. We show that, in certain applications, rightful ownership cannot be resolved by current watermarking schemes alone. Specifically, we attack existing techniques by providing counterfeit watermarking schemes that can be performed on a watermarked image to allow multiple claims of rightful ownership. In the absence of standardization and specific requirements imposed on watermarking procedures, anyone can claim ownership of any watermarked image. In order to protect against the counterfeiting techniques that we develop, we examine the properties necessary for resolving ownership via invisible watermarking. We introduce and study invertibility and quasi-invertibility of invisible watermarking techniques. We propose noninvertible watermarking schemes, and subsequently give examples of techniques that we believe to be nonquasi-invertible and hence invulnerable against more sophisticated attacks proposed in the paper. The attacks and results presented in the paper, and the remedies proposed, further imply that we have to carefully reevaluate the current approaches and techniques in invisible watermarking of digital images based on application domains, and rethink the promises, applications and implications of such digital means of copyright protection.

We shall show that non-invertibility is a necessary but not sufficient condition in resolving ownership disputes. We then define quasi-invertible watermarking schemes, and, present analysis that links invertibility and quasi-invertibility to some classes of watermarking techniques with different properties (which may or may not require original versions in watermark decoding), as well as to the different classes of attacks we have developed.

This paper presents a new criterion for near-lossless image compression and compares it with the more familiar maximum pixel error criterion in terms of compression performance and coded image subjective quality. In this new framework we compress an image in such a way that at each pixel the total error between the original and the coded images over a W/spl times/W window around the pixel does not exceed /spl epsiv/ in magnitude. The new criterion's main advantage is that it preserves image brightness and color. The compression scheme used in this work is similar to a trellis-searched scheme proposed by Ke and Marcellin (1995) in that our scheme is also a predictive, context-based scheme. However, the search for all optimal paths, representing a set of reconstruction values for an image row that satisfies the error criterion and yields the minimum bit rate, has to be done in a planar directed graph instead of a trellis. We show that this framework is also applicable to a joint image filtering and compression scenario. We present simulation results showing the performance of the new error criterion and compare it with the maximum pixel error criterion.

We propose a context-based, adaptive, lossless image codec (CALIC). The codec obtains higher lossless compression of continuous-tone images than other lossless image coding techniques in the literature. This high coding efficiency is accomplished with relatively low time and space complexities. The CALIC puts heavy emphasis on image data modeling. A unique feature of the CALIC is the use of a large number of modeling contexts (states) to condition a nonlinear predictor and adapt the predictor to varying source statistics. The nonlinear predictor can correct itself via an error feedback mechanism by learning from its mistakes under a given context in the past. In this learning process, the CALIC estimates only the expectation of prediction errors conditioned on a large number of different contexts rather than estimating a large number of conditional error probabilities. The former estimation technique can afford a large number of modeling contexts without suffering from the context dilution problem of insufficient counting statistics as in the latter approach, nor from excessive memory use. The low time and space complexities are also attributed to efficient techniques for forming and quantizing modeling contexts.

We propose a context-based, adaptive, lossless image codec (CALIC). CALIC obtains higher lossless compression of continuous-tone images than other techniques reported in the literature. This high coding efficiency is accomplished with relatively low time and space complexities. CALIC puts heavy emphasis on image data modeling. A unique feature of CALIC is the use of a large number of modeling contexts to condition a non-linear predictor and make it adaptive to varying source statistics. The non-linear predictor adapts via an error feedback mechanism. In this adaptation process, CALIC only estimates the expectation of prediction errors conditioned on a large number of contexts rather than estimating a large number of conditional error probabilities. The former estimation technique can afford a large number of modeling contexts without suffering from the sparse context problem. The low time and space complexities of CALIC are attributed to efficient techniques for forming and quantizing modeling contexts.

We investigate lossless compression schemes for video sequences. A simple adaptive prediction scheme is presented that exploits temporal correlations or spectral correlations in addition to spatial correlations. It is seen that even with motion compensation, schemes that utilize only temporal correlations do not perform significantly better than schemes that utilize only spectral correlations. Hence, we look at hybrid schemes that make use of both spectral and temporal correlations. The hybrid schemes give significant improvement in performance over other techniques. Besides prediction schemes, we also look at some simple error modeling techniques that take into account prediction errors made in spectrally and/or temporally adjacent pixels in order to efficiently encode the prediction residual. Implementation results on standard test sequences indicate that significant improvements can be obtained by the proposed techniques.

Conventional vector quantization (VQ)-based techniques partition an image into nonoverlapping blocks that are then raster scanned and quantized. Image blocks that contain an edge result in high-frequency vectors. The coarse representation of such vectors leads to visually annoying degradations in the reconstructed image. The authors present a solution to the edge-degradation problem based on some earlier work on scan models. The approach reduces the number of vectors with abrupt intensity variations by using an appropriate scan to partition an image into vectors. They show how their techniques can be used to enhance the performance of VQ of multispectral data sets. Comparisons with standard techniques are presented and shown to give substantial improvements.

Linear predictive techniques perform poorly when used with color-mapped images where pixel values represent indices that point to color values in a look-up table. Reordering the color table, however, can lead to a lower entropy of prediction errors. In this paper, we investigate the problem of ordering the color table such that the absolute sum of prediction errors is minimized. The problem turns out to be intractable, even for the simple case of one-dimensional (1-D) prediction schemes. We give two heuristic solutions for the problem and use them for ordering the color table prior to encoding the image by lossless predictive techniques. We demonstrate that significant improvements in actual bit rates can be achieved over dictionary-based coding schemes that are commonly employed for color-mapped images.

While there exist many asymmetric techniques for the lossy compression of image data, most techniques reported for lossless compression of image data have been symmetric. In this paper we present a new lossless compression technique that is well suited for asymmetric applications. It gives superior performance compared to other lossless compression techniques reported in the literature. Hence, it can also potentially be adapted for use in symmetric applications that require high compression ratios.

When applying predictive compression on image data there is an implicit assumption that the image is scanned in a particular order. Clearly, depending on the image, a different scanning order may give better compression. In earlier work, we had defined the notion of a prediction tree (or scan) which defines a scanning order for an image. An image can be decorrelated by taking differences among adjacent pixels along any traversal of a scan. Given an image, an optimal scan that minimizes the absolute sum of the differences encountered can be computed efficiently. However, the number of bits required to encode an optimal scan turns out to be prohibitive for most applications. In this paper we present a prediction scheme that partitions an image into blocks and for each block selects a scan from a codebook of scans such that the resulting prediction error is minimized. Techniques based on clustering are developed for the design of a codebook of scans. Design of both semiadaptive and adaptive codebooks is considered. We also combine the new prediction scheme with an effective error modeling scheme. Implementation results are then given, which compare very favorably with the JPEG lossless compression standard.>

We investigate the problem of constructing a prediction scheme for a given image that results in the minimum zero-order entropy of prediction errors. The problem is formulated as a combinatorial optimization problem. This allows the use of some well known techniques from combinatorial optimization in order to construct heuristic solutions. We describe a few heuristics and give preliminary implementation results. The techniques developed can also be generalized in a straight forward manner to composite source modeling where the data is modeled as an interleaved sequence emanating from k different sub-sources. Although the problems and proposed solutions are described in a strictly deterministic manner, they can also be formulated in a stochastic framework to yield solutions that are valid for a family of images emitted by the same source.>

While spatial correlations are adequately exploited by standard lossless image compression techniques, little success has been attained in exploiting spectral correlations when dealing with multispectral image data. The authors present some new lossless image compression techniques that capture spectral correlations as well as spatial correlation in a simple and elegant manner. The schemes are based on the notion of a prediction tree, which defines a noncausal prediction model for an image. The authors present a backward adaptive technique and a forward adaptive technique. They then give a computationally efficient way of approximating the backward adaptive technique. The approximation gives good results and is extremely easy to compute. Simulation results show that for high spectral resolution images, significant savings can be made by using spectral correlations in addition to spatial correlations. Furthermore, the increase in complexity incurred in order to make these gains is minimal.>

Prediction techniques have been applied very successfully in compression of speech data. For image data this paper employs an approach based on spanning trees to construct nonlinear predictive schemes. Images are not scanned in any predetermined fashion, nor is the prediction for any pixel based on a single fixed scheme. Preliminary implementations give promising results over a wide range of images.>