While the Vision Transformer (ViT) architecture gains prominence in computer vision and finds growing applications in edge computing, its lack of strong inductive biases regarding shift, scale, and rotational invariance necessitates pre-training on large-scale datasets. Moreover, the increasing depth and parameter counts in ViT models present significant challenges for training, particularly in edge environments where computational resources are constrained. To mitigate these challenges, this paper introduces a novel Horizontally Scalable Vision Transformer (HSViT) architecture. Specifically, a novel image-level feature embedding approach is introduced that incorporates convolutional layers prior to the Transformer blocks. This design helps preserve inductive biases, allowing the model to potentially eliminate the need for pre-training while achieving strong performance on small datasets. Furthermore, a novel horizontally scalable architecture is designed, facilitating collaborative model training and inference across multiple edge devices. The experimental results show that, without pre-training, HSViT achieves up to 10% higher top-1 accuracy than state-of-the-art methods on several small datasets, while improving the top-1 accuracy of existing CNN backbones by up to 3.1% on ImageNet-1k. The code is available at https://github.com/xuchenhao001/HSViT.

Deepfake detection is essential for mitigating the growing risks associated with the misuse of AI in video manipulation. While existing deep learning-based methods excel in intra-dataset settings, their performance often deteriorates significantly when applied to unseen datasets, underscoring the pressing challenge of improving generalization capability. To tackle this challenge, we focus on identitying discriminative regions by applying the principle of object detection, which locating forged clues. In this paper, we propose an aggregated distinguishable feature learning framework that innovatively incorporates object detector with a graph claasifier. Specifically, the proposed framework consists of three main components: 1) We employ a DEtection TRansformer (DETR) to learn intrinsic feature differences for capturing different discriminative regions. 2) To better utilize distinguishable features in these regions, we treat each local feature vector of the regions as node and design a node correlation generation module (NCGM) to establish the connections between the nodes by integrating the feature similarity and spatial location relationship. 3) we employ graph convolutional neural networks to aggregate the distinguishable features and learn the global connectivity pattern of the nodes for face forgery detection. Comprehensive experimental results on four benchmark datasets demonstrate that our method effectively improves generalization capability and outperforms other state-of-the-art approaches.

Video Instance Segmentation (VIS), also known as multi-object tracking and segmentation, represents a fundamental challenge in computer vision that requires simultaneous detection, segmentation, and tracking of object instances across video frames. This complex task has gained significant attention due to its crucial role in various real-world applications. The advent of deep learning has promoted VIS approaches, leading to numerous architectural innovations and performance improvements. This survey presents a systematic review of deep learning-based VIS methods, introducing a novel categorization based on temporal modeling strategies: frame-by-frame, clip-based, in-memory feature propagation, and in-memory object query propagation. Comprehensive quantitative comparisons of existing work across three major VIS benchmark datasets are also provided. Additionally, emerging challenges in the field are explored, with several promising research directions identified, aiming to provide valuable insights for researchers and practitioners interested in VIS, while further advancing deep learning techniques for VIS. • Categorization of VIS approaches based on their temporal modeling strategies. • Comprehensive quantitative comparison of current VIS methods. • Analysis of the challenges and potential future research directions in VIS.

Generalized age feature extraction is crucial for age-related facial analysis tasks, such as age estimation and age-invariant face recognition (AIFR). Despite the recent successes of models in homogeneous-dataset experiments, their performance drops significantly in cross-dataset evaluations. Most of these models fail to extract generalized age features as they only attempt to map extracted features with training age labels directly without explicitly modeling the natural ordinal progression of aging. In this paper, we propose Order-Enhanced Contrastive Learning (OrdCon), a novel contrastive learning framework designed explicitly for ordinal attributes like age. Specifically, to extract generalized features, OrdCon aligns the direction vector of two features with either the natural aging direction or its reverse to model the ordinal process of aging. To further enhance generalizability, OrdCon leverages a novel soft proxy matching loss as a second contrastive objective, ensuring that features are positioned around the center of each age cluster with minimal intra-class variance and proportionally away from other clusters. By modeling the ageing process, the framework can enhance generalizability by improving the alignment of samples from the same class and reducing the divergence of direction vectors. We demonstrate that our proposed method achieves comparable results to state-of-the-art methods on various benchmark datasets in homogeneous-dataset evaluations for both age estimation and AIFR. In cross-dataset experiments, OrdCon outperforms other methods by reducing the mean absolute error by approximately 1.38 on average for the age estimation task and boosts the average accuracy for AIFR by 1.87%.

The rapid advancement of face manipulation technology has spurred an urgent need for forgery detection. Existing deepfake detection approaches have achieved impressive performance under the intra-dataset scenario where the same algorithm generates training and testing face data. However, the performance is by no means satisfactory when the methods are applied to unseen forgery datasets. To tackle this problem, in this paper, we propose a new perspective of face forgery detection by considering feature inconsistency in spatial and frequency domains in manipulated images. Specifically, we design a two-stream network equipped with a Multi-scale Mutual Local Consistency Learning module (MMLCL) that consists of a Global Enhancement Module (GEM) combining Mutual Local Consistency Learning (MLCL) to learn local consistency in multi-scale enhanced feature maps. We further exploit the mutual representation to obtain an attention map that serves as guidance of forged regions on the output features for final classification. Extensive experiments demonstrate that our proposed method achieves effectiveness and generalization towards unseen face forgeries.

Cross-age facial images are typically challenging and expensive to collect, making noise-free age-oriented datasets relatively small compared to widely-used large-scale facial datasets. Additionally, in real scenarios, images of the same subject at different ages are usually hard or even impossible to obtain. Both of these factors lead to a lack of supervised data, which limits the versatility of supervised methods for age-invariant face recognition, a critical task in applications such as security and biometrics. To address this issue, we propose a novel semi-supervised learning approach named Cross-Age Contrastive Learning (CACon). Thanks to the identity-preserving power of recent face synthesis models, CACon introduces a new contrastive learning method that leverages an additional synthesized sample from the input image. We also propose a new loss function in association with CACon to perform contrastive learning on a triplet of samples. We demonstrate that our method not only achieves state-of-the-art performance in homogeneous-dataset experiments on several age-invariant face recognition benchmarks but also outperforms other methods by a large margin in cross-dataset experiments.

Deep learning has enabled realistic face manipulation for malicious purposes (e.g., deepfakes), which poses significant concerns over the integrity of the media in circulation. Most existing deep learning techniques for deepfake detection can achieve promising performance in the intra-dataset evaluation setting, but are unable to perform satisfactorily in the inter-dataset evaluation setting. Most previous methods use a backbone network to extract global features for making predictions and only employ binary supervision to train the network. Classification merely based on the learning of global features often leads to weak generalizability to deepfakes of unseen manipulation methods. In this paper, we design a two-branch Convolutional AutoEncoder (CAE), which considers the reconstruction and classification tasks simultaneously for deepfake detection. This Joint Reconstruction and Classification (JRC) method shares the information learned by one task with the other, each focusing on different aspects, and hence boosts the overall performance. JRC is end-to-end, and experiments demonstrate that it achieves state-of-the-art performance on three commonly-used datasets, particularly in the cross-dataset evaluation setting.

Deepfake detection plays a key role in preventing the misuses of artificial intelligence in video editing. Current deep learning-based deepfake detection methods often perform quite well in intra-dataset testing, but they may lose good performance in cross-dataset testing. In other words, generalization capability is still a crucial problem to be resolved. In this paper, we address deepfake detection by treating an image as non-Euclidean data and representing it as a graph so as to infer the informative connections between image patches/nodes to improve the detector’s generalization capability. Specifically, we propose a graph neural network-based paradigm that casts deepfake detection as a graph binary classification problem. First, we propose a dual-branch network to extract node features from both RGB images and their color difference images (CDIs) via the Transformer-based trainable node encoder module (TNEM). Second, we adopt the adjacency matrix to establish the connections of the nodes and further optimize the graph representation by applying the adaptive threshold to the adjacency matrix. Third, multi-head graph convolutional neural networks are carried out for node feature extraction. RGB node features and CDI node features are concatenated and separately fed into the graph classifier and node classifier for forgery detection and forgery localization. Experimental results demonstrate that our method can overall outperform other state-of-the-art methods on 7 popular benchmark datasets. Notably, our model achieves the highest AUC values of 96.19%, 80.99% and 87.68% on Celeb-DF-V2, DFDC and DFDCP in turn when trained on FF++ (C23). The visualization of node classification results also provides good interpretability of our proposed approach.

Existing face forgery detection methods often consider the manipulation detection problem as a binary classification problem, which are easily prone to overfitting. The detection performance degrades for datasets not appearing in the training stage. To solve this problem, we utilize the idea of domain generalization and design a hybrid domain meta-learning network (HDMNet) for face forgery detection and manipulation localization. The network allows the reveal of more essential face-swapping traces through multi-domain off-sets. Specifically, HDMNet consists of three modules, namely the feature extraction module (FEM), the face mask representation module (FMRM) and the meta-learning module (MLM). The FEM is composed of a comprehensive feature representation of image color features and high-frequency noise features. The FMRM utilizes graph convolution to predict face masks, providing supervision of domain knowledge. The MLM extracts domain-invariant features through a hybrid domain meta-learning strategy. Extensive experiments on several benchmark databases validate the effectiveness and good generalization ability of our method compared with several state-of-the-art methods.

Resizing input face images of arbitrary sizes to a uniform size is an essential preprocessing to satisfy the architectural requirements of face forgery detectors. In this letter, we reveal an important observation that traditional resizing methods degrade the performance of face forgery detectors due to the loss of high-frequency information. To address this issue, we propose a simple yet effective learnable information-preserving resizer to replace its lossy traditional counterparts. Specifically, we use Haar transform to separate low-and high-frequency components, and then perform learnable resizing on the high-frequency subbands. We conduct experiments to compare our learnable resizer with other methods and evaluate three existing detectors with and without incorporating our resizer. Experimental results show that our resizer outperforms other resizers and consistently enhances the detection performance of tested detectors, confirming the effectiveness of our proposed resizer

Identification of source social networks of images based on the traces left on images by such platforms is a crucial task in image forensics. The existing techniques provide successful solutions to such a problem. However, we show that the state-of-the-art techniques are adversely affected due to the leaking side-channel information from scene details that convolutional neural networks (CNNs) are prone to exploit. Thus, highly correlated scene details in the training and test sets lead to overestimation of the performance. To address this problem, we develop a data-driven system by parallelizing three CNNs having kernels with different sizes that benefit from learning more relevant forensic traces making the model less susceptible to scene content. The experimental results achieved by the proposed model either trained on images with or without scene overlap show that there is no influence of scene content in the feature learning of the proposed method.

With the increasing popularity of convolutional neural networks (CNNs), recent works on face-based age estimation employ these networks as the backbone. However, state-of-the-art CNN-based methods treat each facial region equally, thus entirely ignoring the importance of some facial patches that may contain rich age-specific information. In this paper, we propose a face-based age estimation framework, called Attention-based Dynamic Patch Fusion (ADPF). In ADPF, two separate CNNs are implemented, namely the AttentionNet and the FusionNet. The AttentionNet dynamically locates and ranks age-specific patches by employing a novel Ranking-guided Multi-Head Hybrid Attention (RMHHA) mechanism. The FusionNet uses the discovered patches along with the facial image to predict the age of the subject. Since the proposed RMHHA mechanism ranks the discovered patches based on their importance, the length of the learning path of each patch in the FusionNet is proportional to the amount of information it carries (the longer, the more important). ADPF also introduces a novel diversity loss to guide the training of the AttentionNet and reduce the overlap among patches so that the diverse and important patches are discovered. Through extensive experiments, we show that our proposed framework outperforms state-of-the-art methods on several age estimation benchmark datasets.

Abstract In this paper, we investigate the influential factors that impact on the performance when the tasks are co-running on a multicore computers. Further, we propose the machine learning-based prediction framework to predict the performance of the co-running tasks. In particular, two prediction frameworks are developed for two types of task in our model: repetitive tasks (i.e., the tasks that arrive at the system repetitively) and new tasks (i.e., the task that are submitted to the system the first time). The difference between which is that we have the historical running information of the repetitive tasks while we do not have the prior knowledge about new tasks. Given the limited information of the new tasks, an online prediction framework is developed to predict the performance of co-running new tasks by sampling the performance events on the fly for a short period and then feeding the sampled results to the prediction framework. We conducted extensive experiments with the SPEC2006 benchmark suite to compare the effectiveness of different machine learning methods considered in this paper. The results show that our prediction model can achieve the accuracy of 99.38% and 87.18% for repetitive tasks and new tasks, respectively.

Localizing image forgeries is one of the key topics in multimedia forensics. Among many image forgery localization techniques, the one based on the photo-response non-uniformity (PRNU) noise has attracted substantial attention because of its capability of localizing forgeries regardless of the type of forgery. However, despite the devoted efforts to improving the performance of PRNU-based forgery localization, there remain challenges to be overcome, especially for detecting subtle forgeries in PRNU-attenuated regions due to complex image content. In this work, we investigate the feasibility and effectiveness of convolutional neural networks (CNN) in predicting PRNU correlations under complex backgrounds for more accurate forgery localization. The experimental results on 20 cameras and 200 realistic forgery images show that significant improvement in correlation prediction and forgery localization can be achieved even with a light-weight CNN model. The robustness of different correlation predictors against JPEG compression is also evaluated.

Face deepfake detection has seen impressive results recently. Nearly all existing deep learning techniques for face deepfake detection are fully supervised and require labels during training. In this paper, we design a novel deepfake detection method via unsupervised contrastive learning. We first generate two different transformed versions of an image and feed them into two sequential sub-networks, i.e., an encoder and a projection head. The unsupervised training is achieved by maximizing the correspondence degree of the outputs of the projection head. To evaluate the detection performance of our unsupervised method, we further use the unsupervised features to train an efficient linear classification network. Extensive experiments show that our unsupervised learning method enables comparable detection performance to state-of-the-art supervised techniques, in both the intra- and inter-dataset settings. We also conduct ablation studies for our method.

Photo Response Non-Uniformity (PRNU) has been used as a powerful device fingerprint for image forgery detection because image forgeries can be revealed by finding the absence of the PRNU in the manipulated areas. The correlation between an image's noise residual with the device's reference PRNU is often compared with a decision threshold to check the existence of the PRNU. A PRNU correlation predictor is usually used to determine this decision threshold assuming the correlation is content-dependent. However, we found that not only the correlation is content-dependent, but it also depends on the camera sensitivity setting. Camera sensitivity, commonly known by the name of ISO speed, is an important attribute in digital photography. In this work, we will show the PRNU correlation's dependency on ISO speed. Due to such dependency, we postulate that a correlation predictor is ISO speed-specific, i.e. reliable correlation predictions can only be made when a correlation predictor is trained with images of similar ISO speeds to the image in question. We report the experiments we conducted to validate the postulate. It is realized that in the real-world, information about the ISO speed may not be available in the metadata to facilitate the implementation of our postulate in the correlation prediction process. We hence propose a method called Content-based Inference of ISO Speeds (CINFISOS) to infer the ISO speed from the image content.

With the assistance of sophisticated training methods applied to single labeled datasets, the performance of fully-supervised person re-identification (Person Re-ID) has been improved significantly in recent years. However, these models trained on a single dataset usually suffer from considerable performance degradation when applied to videos of a different camera network. To make Person Re-ID systems more practical and scalable, several cross-dataset domain adaptation methods have been proposed, which achieve high performance without the labeled data from the target domain. However, these approaches still require the unlabeled data of the target domain during the training process, making them impractical. A practical Person Re-ID system pre-trained on other datasets should start running immediately after deployment on a new site without having to wait until sufficient images or videos are collected and the pre-trained model is tuned. To serve this purpose, in this paper, we reformulate person re-identification as a multi-dataset domain generalization problem. We propose a multi-dataset feature generalization network (MMFA-AAE), which is capable of learning a universal domain-invariant feature representation from multiple labeled datasets and generalizing it to 'unseen' camera systems. The network is based on an adversarial auto-encoder to learn a generalized domain-invariant latent feature representation with the Maximum Mean Discrepancy (MMD) measure to align the distributions across multiple domains. Extensive experiments demonstrate the effectiveness of the proposed method. Our MMFA-AAE approach not only outperforms most of the domain generalization Person Re-ID methods, but also surpasses many state-of-the-art supervised methods and unsupervised domain adaptation methods by a large margin.

The vanilla Generative Adversarial Networks (GANs) are commonly used to generate realistic images depicting aged and rejuvenated faces. However, the performance of such vanilla GANs in the age-oriented face synthesis task is often compromised by the mode collapse issue, which may produce poorly synthesized faces with indistinguishable visual variations. In addition, recent age-oriented face synthesis methods use the L1 or L2 constraint to preserve the identity information in synthesized faces, which implicitly limits the identity permanence capabilities when these constraints are associated with a trivial weighting factor. In this paper, we propose a method for the age-oriented face synthesis task that achieves high synthesis accuracy with strong identity permanence capabilities. Specifically, to achieve high synthesis accuracy, our method tackles the mode collapse issue with a novel Conditional Discriminator Pool, which consists of multiple discriminators, each targeting one particular age category. To achieve strong identity permanence capabilities, our method uses a novel Adversarial Triplet loss. This loss, which is based on the Triplet loss, adds a ranking operation to further pull the positive embedding towards the anchor embedding to significantly reduce intra-class variances in the feature space. Through extensive experiments, we show that our proposed method outperforms state-of-the-art methods in terms of synthesis accuracy and identity permanence capabilities, both qualitatively and quantitatively.

Understanding human actions from videos captured by drones is a challenging task in computer vision due to the unfamiliar viewpoints of individuals and changes in their size due to the camera's location and motion. This work proposes DroneCaps, a capsule network architecture for multi-label human action recognition (HAR) in videos captured by drones. DroneCaps uses features computed by 3D convolution neural networks plus a new set of features computed by a novel Binary Volume Comparison layer. All these features, in conjunction with the learning power of CapsNets, allow understanding and abstracting the different viewpoints and poses of the depicted individuals very efficiently, thus improving multi-label HAR. The evaluation of the DroneCaps architecture's performance for multi-label classification shows that it outperforms state-of-the-art methods on the Okutama-Action dataset.

Device fingerprints like sensor pattern noise (SPN) are widely used for provenance analysis and image authentication. Over the past few years, the rapid advancement in digital photography has greatly reshaped the pipeline of image capturing process on consumer-level mobile devices. The flexibility of camera parameter settings and the emergence of multi-frame photography algorithms, especially high dynamic range (HDR) imaging, bring new challenges to device fingerprinting. The subsequent study on these topics requires a new purposefully built image dataset. In this paper, we present the Warwick Image Forensics Dataset, an image dataset of more than 58,600 images captured using 14 digital cameras with various exposure settings. Special attention to the exposure settings allows the images to be adopted by different multi-frame computational photography algorithms and for subsequent device fingerprinting. The dataset is released as an open-source, free for use for the digital forensic community.

Capsule Networks (CapsNets) are recently introduced to overcome some of the shortcomings of traditional Convolutional Neural Networks (CNNs). CapsNets replace neurons in CNNs with vectors to retain spatial relationships among the features. In this paper, we propose a CapsNet architecture that employs individual video frames for human action recognition without explicitly extracting motion information. We also propose weight pooling to reduce the computational complexity and improve the classification accuracy by appropriately removing some of the extracted features. We show how the capsules of the proposed architecture can encode temporal information by using the spatial features extracted from several video frames. Compared with a traditional CNN of the same complexity, the proposed CapsNet improves action recognition performance by 12.11% and 22.29% on the KTH and UCF-sports datasets, respectively.

In this paper, we propose a compact and low-complexity binary feature descriptor for video analytics. Our binary descriptor encodes the motion information of a spatio-temporal support region into a low-dimensional binary string. The descriptor is based on a binning strategy and a construction that binarizes separately the horizontal and vertical motion components of the spatio-temporal support region. We pair our descriptor with a novel Fisher Vector (FV) scheme for binary data to project a set of binary features into a fixed length vector in order to evaluate the similarity between feature sets. We test the effectiveness of our binary feature descriptor with FVs for action recognition, which is one of the most challenging tasks in computer vision, as well as gait recognition and animal behavior clustering. Several experiments on the KTH, UCF50, UCF101, CASIA-B, and TIGdog datasets show that the proposed binary feature descriptor outperforms the state-of-the-art feature descriptors in terms of computational time and memory and storage requirements. When paired with FVs, the proposed feature descriptor attains a very competitive performance, outperforming several state-of-the-art feature descriptors and some methods based on convolutional neural networks.

Most existing source-oriented image and video clustering algorithms based on sensor pattern noise (SPN) rely on the pairwise similarities, whose calculation usually dominates the overall computational time. The heavy computational burden is mainly incurred by the high dimensionality of SPN, which typically goes up to millions for delivering plausible clustering performance. This problem can be further aggravated by the uncertainty of the orientation of images or videos because the spatial correspondence between data with uncertain orientations needs to be reestablished in a brute-force search manner. In this work, we propose a rotation-invariant binary representation of SPN to address the issue of rotation and reduce the computational cost of calculating the pairwise similarities. Results on two public multimedia forensics databases have shown that the proposed approach is effective in overcoming the rotation issue and speeding up the calculation of pairwise SPN similarities for source-oriented image and video clustering.

Millions of surveillance cameras are currently installed in public places around the world, making it necessary to intelligently analyse the acquired data to detect the occurrence of abnormal events. A vast number of methods to detect such events have been recently proposed; unfortunately, there is a lack of methods capable of detecting these events as frames are acquired, also known as online processing. In this paper, we present an online framework for video anomaly detection that employs binary features to encode motion information, and low-complexity probabilistic models for detection. Evaluation results on the popular UCSD dataset and on a recently introduced real-event video surveillance dataset show that our framework outperforms non-online and online methods.

Nowadays, big imaging data are very common in many fields of study. As a result, detecting small objects in very large images is challenging and computationally demanding. Taking advantage of the intrinsic cumulative properties of the Fisher Score, we propose the Integral Fisher Score (IFS) for low-complexity and accurate object detection in big imaging data. The IFS, which is a multi-dimensional extension of the Integral Image, allows computing the Fisher Vector associated with a spatial region using only four operations. This considerably reduces the computational cost of searching for a small query object on a very large target image. Evaluations for the detection of small object on high-resolution HUB telescope and digital pathology images show that IFS attains a high accuracy with short processing times.

Convolutional Neural Networks (CNN) have been applied to age-related research as the core framework. Although faces are composed of numerous facial attributes, most works with CNNs still consider a face as a typical object and do not pay enough attention to facial regions that carry age-specific feature for this particular task. In this paper, we propose a novel CNN architecture called Fusion Network (Fusion-Net) to tackle the age estimation problem. Apart from the whole face image, the FusionNet successively takes several age-specific facial patches as part of the input to emphasize the age-specific features. Through experiments, we show that the FusionNet significantly outperforms other state-of-the-art models on the MORPH II benchmark.

Sensor pattern noise (SPN) is an inherent fingerprint of imaging devices, which provides an effective way for source camera identification (SCI). Although SPNs extracted from large image blocks usually yield high identification accuracy, their high dimensionality would incur a high computational cost in the matching stage, consequently hindering many applications that require efficient camera matchings. In this work, we employ and evaluate the concept of principal component analysis (PCA) de-noising in SCI tasks. Based on this concept, we present a framework that formulates a compact SPN representation. To enhance the de-noising effect, we introduce a training set construction procedure that minimizes the impact of various interfering artifacts, which is especially useful in some challenging cases, e.g., when only textured reference images are available. To further boost the SCI performance, a novel approach based on linear discriminant analysis (LDA) is adopted to extract more discriminant SPN features. To evaluate our methods, extensive experiments are conducted on the Dresden image database. The results indicate that the proposed framework can serve as an effective post-processing procedure, which not only boosts the performance, but also greatly reduces the computational cost in the matching phase.

The Pap test is a manual screening procedure that is used to detect the precursor lesions of cervical cancer by analyzing changes in nuclei and cytoplasms of cervical cells. Due to the sensitivity of the Pap test to intra- and inter-observer variability, automating the procedure using digital image analysis test is still an open problem. Within this context, segmentation of overlapping cervical cells is a key component to develop image analysis methods. In this paper, we propose a framework for segmenting the cytoplasm of each individual cell depicted within an image of overlapping cervical cells. The proposed framework uses a patch-based approach where a parametric active contour detects, on a patch-by-path basis, the cytoplasm boundary of each overlapping cell. The active contour within the patch deforms under the influence of Gradient Vector Flow (GVF) forces computed based on the local edges depicted in each patch region. Results show that the proposed framework achieves more accurate cytoplasm segmentation results compared to the current state-of-art methods.

Gait recognition can be performed without subject cooperation under harsh conditions, thus it is an important tool in forensic gait analysis, security control, and other commercial applications. One critical issue that prevents gait recognition systems from being widely accepted is the performance drop when the camera viewpoint varies between the registered templates and the query data. In this paper, we explore the potential of combining feature optimisers and representations learned by convolutional neural networks (CNN) to achieve efficient view-invariant gait recognition. The experimental results indicate that CNN learns highly discriminative representations across moderate view variations, and these representations can be further improved using view-invariant feature selectors, achieving a high matching accuracy across views.

In this paper, we show how a dealer with limited resources is possible to share the secrets to players via an untrusted cloud server without compromising the privacy of the secrets. This scheme permits a batch of two secret messages to be shared to two players in such a way that the secrets are reconstructable if and only if two of them collaborate. An individual share reveals absolutely no information about the secrets to the player. The secret messages are obfuscated by encryption and thus give no information to the cloud server. Furthermore, the scheme is compatible with the Paillier cryptosystem and other cryptosystems of the same type. In light of the recent developments in privacy-preserving watermarking technology, we further model the proposed scheme as a variant of reversible watermarking in the encrypted domain.

A hierarchical relaxed partitioning system (HRPS) is proposed for recognizing similar activities which has a feature space with multiple overlaps. Two feature descriptors are built from the human motion analysis of a 2-D stick figure to represent cyclic and noncyclic activities. The HRPS first discerns the pure and impure activities, i.e., with no overlaps and multiple overlaps in the feature space, respectively, then tackles the multiple overlaps problem of the impure activities via an innovative majority voting scheme. The results show that the proposed method robustly recognizes various activities of two different resolution data sets, i.e., low and high (with different views). The advantage of HRPS lies in the real-time speed, ease of implementation and extension, and nonintensive training.

Practical applications of digital forensics are often faced with the challenge of grouping large-scale suspicious images into a vast number of clusters, each containing images taken by the same camera. This task can be approached by resorting to the use of sensor pattern noise (SPN), which serves as the fingerprint of the camera. The challenges of large-scale image clustering come from the sheer volume of the image set and the high dimensionality of each image. The difficulties can be further aggravated when the number of classes (i.e., the number of cameras) is much higher than the average size of class (i.e., the number of images acquired by each camera). We refer to this as the NC ≫ SC problem, which is not uncommon in many practical scenarios. In this paper, we propose a novel clustering framework that is capable of addressing the NC ≫ SC problem without a training process. The proposed clustering framework was evaluated on the Dresden image database and compared with the state-of-the-art SPN-based image clustering algorithms. Experimental results show that the proposed clustering framework is much faster than the state-of-the-art algorithms while maintaining a high level of clustering quality.

Level set methods have been widely used to implement active contours for image segmentation applications due to their good boundary detection accuracy. In the context of medical image segmentation, weak edges and inhomogeneities remain important issues that may hinder the accuracy of any segmentation method based on active contours implemented using level set methods. This paper proposes a method based on active contours implemented using level set methods for segmentation of such medical images. The proposed method uses a level set evolution that is based on the minimization of an objective energy functional whose energy terms are weighted according to their relative importance in detecting boundaries. This relative importance is computed based on local edge features collected from the adjacent region located inside and outside of the evolving contour. The local edge features employed are the edge intensity and the degree of alignment between the image's gradient vector flow field and the evolving contour's normal. We evaluate the proposed method for segmentation of various regions in real MRI and CT slices, X-ray images, and ultra sound images. Evaluation results confirm the advantage of weighting energy forces using local edge features to reduce leakage. These results also show that the proposed method leads to more accurate boundary detection results than state-of-the-art edge-based level set segmentation methods, particularly around weak edges.

Over the past decade, video anomaly detection has been explored with remarkable results. However, research on methodologies suitable for online performance is still very limited. In this paper, we present an online framework for video anomaly detection. The key aspect of our framework is a compact set of highly descriptive features, which is extracted from a novel cell structure that helps to define support regions in a coarse-to-fine fashion. Based on the scene's activity, only a limited number of support regions are processed, thus limiting the size of the feature set. Specifically, we use foreground occupancy and optical flow features. The framework uses an inference mechanism that evaluates the compact feature set via Gaussian Mixture Models, Markov Chains, and Bag-of-Words in order to detect abnormal events. Our framework also considers the joint response of the models in the local spatio-temporal neighborhood to increase detection accuracy. We test our framework on popular existing data sets and on a new data set comprising a wide variety of realistic videos captured by surveillance cameras. This particular data set includes surveillance videos depicting criminal activities, car accidents, and other dangerous situations. Evaluation results show that our framework outperforms other online methods and attains a very competitive detection performance compared with state-of-the-art non-online methods.

Inspired by the binary-based descriptors (e.g. LBP, ALOHA, FREAK, BRISK), we propose the 3D Binary Pair Differences (3DBPD) video descriptor for action recognition. By comparing several spatio-temporal sub-regions around interests points, our descriptor is a feature vector with a dimensionality of up to 30% smaller than that of existing state-of-the-art descriptors. We demonstrate the effectiveness of the 3DBPD descriptor for action recognition with a SVM classifier and a simple Bag Of Video Words (BOV) generated using k-means. The proposed descriptor has very competitive recognition rates compared to other state-of-the-art descriptors, with an outstanding performance in terms of speed. Additionally, the 3DBPD descriptor requires a small codebook compared to those required by existing BOV-based descriptors.

In this work, we propose a method to obtain higher quality sensor pattern noise (SPN) for identifying source cameras. We believe that some components of SPN have been severely contaminated by the errors introduced by denoising filters and the quality of SPN can be improved by abandoning those components. In our proposed method, some coefficients with higher denoising errors are abandoned in the wavelet representation of SPN and the remaining wavelet coefficients are further enhanced to suppress the scene details in the SPN. These two steps aim to provide better SPN with higher signal-to-noise ratio (SNR) and therefore improve the identification performance. The experimental results on 2,000 images captured by 10 cameras (each responsible for 200 images), show that our method achieves better receiver operating characteristic (ROC) performance when compared with some state-of-the-art methods.

Although sensor pattern noise (SPN) has been proved to be an effective means to uniquely identify digital cameras, some non-unique artifacts, shared among cameras undergo the same or similar in-camera processing procedures, often give rise to false identifications. Therefore, it is desirable and necessary to suppress these unwanted artifacts so as to improve the accuracy and reliability. In this paper, we propose a novel preprocessing approach for attenuating the influence of the non-unique artifacts on the reference SPN to reduce the false identification rate. Specifically, we equalize the magnitude spectrum of the reference SPN through detecting and suppressing the peaks according to the local characteristics, aiming at removing the interfering periodic artifacts. Combined with six SPN extractions or enhancement methods, our proposed spectrum equalization algorithm is evaluated on the Dresden image database as well as our own database, and compared with the state-of-the-art preprocessing schemes. The experimental results indicate that the proposed procedure outperforms, or at least performs comparable with, the existing methods in terms of the overall receiver operating characteristic curves and kappa statistic computed from a confusion matrix, and tends to be more resistant to JPEG compression for medium and small image blocks.

Sensor Pattern Noise (SPN) has been proved as an effective fingerprint of imaging devices to link pictures to the cameras that acquired them. In practice, forensic investigators usually extract this camera fingerprint from large image block to improve the matching accuracy because large image blocks tend to contain more SPN information. As a result, camera fingerprints usually have a very high dimensionality. However, the high dimensionality of fingerprint will incur a costly computation in the matching phase, thus hindering many interesting applications which require an efficient real-time camera matching. To solve this problem, an effective feature extraction method based on PCA and LDA is proposed in this work to compress the dimensionality of camera fingerprint. Our experimental results show that the proposed feature extraction algorithm could greatly reduce the size of fingerprint and enhance the performance in term of Receiver Operating Characteristic (ROC) curve of several existing methods.

Sensor Pattern Noise (SPN) is an inherent fingerprint of imaging devices, which has been widely used in the tasks of digital camera identification, image classification and forgery detection. In our previous work, a feature extraction method based on PCA denoising concept was applied to extract a set of principal components from the original noise residual. However, this algorithm is inefficient when query cameras are continuously received. To solve this problem, we propose an extension based on Candid Covariance-free Incremental PCA (CCIPCA) and two modifications to incrementally update the feature extractor according to the received cameras. Experimental results show that the PCA and CCIPCA based features both outperform their original features on the ROC performance, and CCIPCA is more efficient on camera updating.

Pure appearance based Hand Gesture Recognition and Spotting in uncontrolled environments are challenging tasks due to the uncontrolled scene settings include: multiple hand regions in the scene; background moving objects; scale, speed and location variations of the gesture trajectories; changing lighting conditions and frontal occlusions. An appearance based method based on a novel classifier weighting scheme is proposed in this paper for hand gesture recognition and spotting in uncontrolled environments. The method is capable of producing decent performance with the presence of all the aforementioned challenges. Two databases are used for evaluating the proposed method, the Palm Graffiti Digits Database and the Warwick Hand Gesture Database. The experimental results demonstrate that the proposed method can deal with the challenges from uncontrolled environments without any prior knowledge and enhance the performance of the initial classifier.

Fast camera fingerprint search is an important issue for source camera identification in real-world applications. So far there has been little work done in this area. In this paper, we propose a novel fast search algorithm. We use global information derived from the relationship between the query fingerprint/digest and the reference fingerprints/digests in the database to guide fast search. This information can provide more accurate and robust clues for the selection of candidate matching database fingerprints. Because the quality of query fingerprints may degrade or vary in realistic applications, the construction of robust search clues is significant. To speed up the search process, we adopt a lookup table that is built on the separate-chaining hash table. The proposed algorithm has been tested using query images from real-world photos. Experiments demonstrate that our algorithm can well adapt to query fingerprints with different quality. It can achieve higher detection rates with lower computational cost than the traditional brute-force search algorithm and a pioneering fast search algorithm in literature.

Robust human gait recognition is challenging because of the presence of covariate factors such as carrying condition, clothing, walking surface, etc. In this paper, we model the effect of covariates as an unknown partial feature corruption problem. Since the locations of corruptions may differ for different query gaits, relevant features may become irrelevant when walking condition changes. In this case, it is difficult to train one fixed classifier that is robust to a large number of different covariates. To tackle this problem, we propose a classifier ensemble method based on the random subspace Method (RSM) and majority voting (MV). Its theoretical basis suggests it is insensitive to locations of corrupted features, and thus can generalize well to a large number of covariates. We also extend this method by proposing two strategies, i.e, local enhancing (LE) and hybrid decision-level fusion (HDF) to suppress the ratio of false votes to true votes (before MV). The performance of our approach is competitive against the most challenging covariates like clothing, walking surface, and elapsed time. We evaluate our method on the USF dataset and OU-ISIR-B dataset, and it has much higher performance than other state-of-the-art algorithms.

Active contours, or snakes, have been widely used for image segmentation purposes. However, high noise sensitivity and poor performance over weak edges are the most acute issues that hinder the segmentation accuracy of these curves, particularly in medical images. In order to overcome these issues, we propose a novel external force that integrates gradient vector flow (GVF) field forces and balloon forces based on a weighting factor computed according to local image features. The proposed external force reduces noise sensitivity, improves performance over weak edges and allows initialization with a single manually selected point. We evaluate the proposed external force for segmentation of various regions on real MRI and CT slices. Evaluation results show that the proposed approach leads to more accurate segmentation than snakes using traditional external forces.

Face recognition (FR) systems in real-world applications need to deal with a wide range of interferences, such as occlusions and disguises in face images. Compared with other forms of interferences such as nonuniform illumination and pose changes, face with occlusions has not attracted enough attention yet. A novel approach, coined dynamic image-to-class warping (DICW), is proposed in this work to deal with this challenge in FR. The face consists of the forehead, eyes, nose, mouth, and chin in a natural order and this order does not change despite occlusions. Thus, a face image is partitioned into patches, which are then concatenated in the raster scan order to form an ordered sequence. Considering this order information, DICW computes the image-to-class distance between a query face and those of an enrolled subject by finding the optimal alignment between the query sequence and all sequences of that subject along both the time dimension and within-class dimension. Unlike most existing methods, our method is able to deal with occlusions which exist in both gallery and probe images. Extensive experiments on public face databases with various types of occlusions have confirmed the effectiveness of the proposed method.

In this paper, a novel forensic method of exposing cut-and-paste image forgery through detecting contrast enhancement is proposed. We reveal the inter-channel correlation introduced by color image interpolation, and show how a linear or nonlinear contrast enhancement can disturb this natural inter-channel dependency. We then construct a metric to measure these correlations, which are useful in distinguishing the original and contrast enhanced images. The effectiveness of the proposed algorithm is experimentally validated on natural color images captured by commercial cameras. Finally, its robustness against some anti-forensic algorithms is also discussed.

The main obstructions of making hand gesture recognition methods robust in real-world applications are the challenges from the uncontrolled environments, including: gesturing hand out of the scene, pause during gestures, complex background, skin-coloured regions moving in background, performers wearing short sleeve and face overlapping with hand. Therefore, a framework for real-time hand gesture recognition in uncontrolled environments is proposed in this paper. A novel tracking scheme is proposed to track multiple hand candidates in unconstrained background, and a weighting model for gesture classification based on Hidden Conditional Random Fields which takes trajectories of multiple hand candidates under different frame rates into consideration is also introduced. The framework achieved invariance under change of scale, speed and location of the hand gestures. The Experimental results of the proposed framework on Palm Graffiti Digits database and Warwick Hand Gesture database show that it can perform well in uncontrolled environments.

This paper presents a new fragile watermarking method, whereby two mechanisms are hierarchically structured to provide self-recovery capabilities. The first one is a secure block-wise mechanism, resilient to cropping, aimed at localising altered pixel-blocks. The second one is an iterative mechanism capable of reconstructing the original contents, by means of exhaustive attempts. The key features of the proposed method, which compare favourably to those of existing schemes, are low embedding distortion and resilience to cropping. Results demonstrate that the proposed scheme is capable of restoring the altered contents, even when the tampered region covers up to 32% of the total pixels in the image.

The last few years have seen the use of photo response non-uniformity noise (PRNU), a unique fingerprint of imaging sensors, in various digital forensic applications such as source device identification, content integrity verification, and authentication. However, the use of a color filter array for capturing only one of the three color components per pixel introduces color interpolation noise, while the existing methods for extracting PRNU provide no effective means for addressing this issue. Because the artificial colors obtained through the color interpolation process are not directly acquired from the scene by physical hardware, we expect that the PRNU extracted from the physical components, which are free from interpolation noise, should be more reliable than that from the artificial channels, which carry interpolation noise. Based on this assumption we propose a couple-decoupled PRNU (CD-PRNU) extraction method, which first decomposes each color channel into four sub-images and then extracts the PRNU noise from each sub-image. The PRNU noise patterns of the sub-images are then assembled to get the CD-PRNU. This new method can prevent the interpolation noise from propagating into the physical components, thus improving the accuracy of device identification and image content integrity verification.

The imaging sensor (e.g., CCD) pattern noise is a noise-like spread-spectrum signal inherently cast onto every digital image by each imaging device. It has been recognized as a reliable device fingerprint for source camera/scanner identification and image origin verification. However, one weakness of current sensor pattern noise-based camera fingerprints is that their construction only uses information from single color channel images. The resultant fingerprint can not comprehensively reflect characteristics of the camera sensor array. Taking into account the characteristics of the CFA (color filter array) structure, this work proposes a new way to construct camera fingerprints using information from all of the three single color channel images. Good detection results have been achieved on different camera models.

We present in this work an unsupervised image classifier, which is capable of clustering images taken by an unknown number of unknown digital cameras into a number of classes, each corresponding to one camera. The classification system first extracts and enhances a sensor pattern noise (SPN) from each image, which serves as the fingerprint of the camera that has taken the image. Secondly, it applies an unsupervised classifier trainer to a small training set of randomly selected SPNs to cluster the SPNs into classes and uses the centroids of those identified classes as the trained classifier. The classifier trainer treats each SPN as a random variable and uses Markov random field (MRF) approach to iteratively assigns a class label to each SPN (i.e., random variable) based on the class labels assigned to the members of a small set of SPNs, called membership committee, and the similarity values between it and the members of the membership committee until a stop criteria is met. The classifier trainer requires no a priori knowledge about the dataset from the user. Finally the image not included in the small training set are classified using the trained classifier depending on the similarity between their SPNs and the centroids of the trained classifier.

The last few years have seen the use of photo response non-uniformity noise (PRNU) - a unique fingerprint of imaging sensors, in various digital forensic applications such as source device identification, content integrity verification and authentication. However, the use of a colour filter array for capturing only one of the three colour components per pixel introduces colour interpolation error, while the existing methods for extracting PRNU do not take this into account and include the colour interpolation error as part of the PRNU, which leaves room for improvement. In this work we propose a new way of extracting PRNU, called Colour-Decoupled PRNU (CD-PRNU), by exploiting the difference between the physical and artificial colour components of the photos taken by digital cameras that use a colour filter array for interpolating artificial colour components from the physical ones. Experimental results presented in this work have shown the superiority of the proposed CD-PRNU to the commonly used version.

BACKGROUND: Time-course microarray experiments can produce useful data which can help in understanding the underlying dynamics of the system. Clustering is an important stage in microarray data analysis where the data is grouped together according to certain characteristics. The majority of clustering techniques are based on distance or visual similarity measures which may not be suitable for clustering of temporal microarray data where the sequential nature of time is important. We present a Granger causality based technique to cluster temporal microarray gene expression data, which measures the interdependence between two time-series by statistically testing if one time-series can be used for forecasting the other time-series or not. RESULTS: A gene-association matrix is constructed by testing temporal relationships between pairs of genes using the Granger causality test. The association matrix is further analyzed using a graph-theoretic technique to detect highly connected components representing interesting biological modules. We test our approach on synthesized datasets and real biological datasets obtained for Arabidopsis thaliana. We show the effectiveness of our approach by analyzing the results using the existing biological literature. We also report interesting structural properties of the association network commonly desired in any biological system. CONCLUSIONS: Our experiments on synthesized and real microarray datasets show that our approach produces encouraging results. The method is simple in implementation and is statistically traceable at each step. The method can produce sets of functionally related genes which can be further used for reverse-engineering of gene circuits.

Sensor pattern noises (SPNs), extracted from digital images to serve as the fingerprints of imaging devices, have been proved as an effective way for digital device identification. However, as we demonstrate in this work, the limitation of the current method of extracting SPNs is that the SPNs extracted from images can be severely contaminated by details from scenes, and as a result, the identification rate is unsatisfactory unless images of a large size are used. In this work, we propose a novel approach for attenuating the influence of details from scenes on SPNs so as to improve the device identification rate of the identifier. The hypothesis underlying our SPN enhancement method is that the stronger a signal component in an SPN is, the less trustworthy the component should be, and thus should be attenuated. This hypothesis suggests that an enhanced SPN can be obtained by assigning weighting factors inversely proportional to the magnitude of the SPN components.

We present the first report on Granger causality based detection of functional modules from temporal gene expression data. The approach uses temporal causal relationships shared between pair of genes to derive a connection matrix, which is further analyzed using graph-theoretic techniques. The approach is evaluated against a synthesized dataset and a real biological dataset obtained for Arabidopsis thaliana. We show the effectiveness of our approach by analyzing the results using the existing biological literature.

Various multivariate time series analysis techniques have been developed with the aim of inferring causal relations between time series. Previously, these techniques have proved their effectiveness on economic and neurophysiological data, which normally consist of hundreds of samples. However, in their applications to gene regulatory inference, the small sample size of gene expression time series poses an obstacle. In this paper, we describe some of the most commonly used multivariate inference techniques and show the potential challenge related to gene expression analysis. In response, we propose a directed partial correlation (DPC) algorithm as an efficient and effective solution to causal/regulatory relations inference on small sample gene expression data. Comparative evaluations on the existing techniques and the proposed method are presented. To draw reliable conclusions, a comprehensive benchmarking on data sets of various setups is essential. Three experiments are designed to assess these methods in a coherent manner. Detailed analysis of experimental results not only reveals good accuracy of the proposed DPC method in large-scale prediction, but also gives much insight into all methods under evaluation.

Sensor pattern noises (SPN), extracted from digital images as device fingerprints, have been proved as an effective way for digital device identification. However, the limitation of the current method of extracting the sensor pattern noise is that the SPNs extracted from images are highly contaminated by the details from the scene and as a result the misclassification rate is high unless images of large size are used. In this work we propose a novel approach for enhancing sensor pattern noises so as to improve the performance of the identifier. The hypothesis underlying our fingerprint enhancer is that the stronger a signal component is, the less trustworthy the component should be and thus should be attenuated. An enhanced fingerprint can be obtained by assigning weighting factors inversely proportional to the magnitude of the signal components.

A reversible (also called lossless, distortion-free, or invertible) visible watermarking scheme is proposed to satisfy the applications, in which the visible watermark is expected to combat copyright piracy but can be removed to losslessly recover the original image. We transparently reveal the watermark image by overlapping it on a user-specified region of the host image through adaptively adjusting the pixel values beneath the watermark, depending on the human visual system-based scaling factors. In order to achieve reversibility, a reconstruction/recovery packet, which is utilized to restore the watermarked area, is reversibly inserted into non-visibly-watermarked region. The packet is established according to the difference image between the original image and its approximate version instead of its visibly watermarked version so as to alleviate its overhead. For the generation of the approximation, we develop a simple prediction technique that makes use of the unaltered neighboring pixels as auxiliary information. The recovery packet is uniquely encoded before hiding so that the original watermark pattern can be reconstructed based on the encoded packet. In this way, the image recovery process is carried out without needing the availability of the watermark. In addition, our method adopts data compression for further reduction in the recovery packet size and improvement in embedding capacity. The experimental results demonstrate the superiority of the proposed scheme compared to the existing methods.

Based on the correlation between expression and ontology-driven gene similarity, we incorporate functional annotations into gene expression clustering validation. A probabilistic framework is proposed to accommodate incomplete annotations, after establishing a new term-term distance measure based on graph theory. Comprehensive evaluations are performed on six clustering algorithms. This study is the first to explore a robust quantitative functional relationship between clusters of genes. Such indices assess clustering quality in terms of consistency of annotation information and serve as new tools for combining biological knowledge with experimental data.

In this paper, we propose two rotation and scale invariant features extracted from the Hough transform domain to guide a CBIR system in the search of relevant building images. Upon receiving a query image, the CBIR system transforms the edges detected from the query into the Hough domain with 180 degrees/bins. From each bin, the peak percentage and peak distance ratio are calculated. The circular correlations between the peak percentages and peak distance ratios across the 180 bins of the query image and those of the database images are then taken as the similarity measure for ranking the relevance of the database images to the query.

MOTIVATION: There is a growing interest in extracting statistical patterns from gene expression time-series data, in which a key challenge is the development of stable and accurate probabilistic models. Currently popular models, however, would be computationally prohibitive unless some independence assumptions are made to describe large-scale data. We propose an unsupervised conditional random fields (CRF) model to overcome this problem by progressively infusing information into the labelling process through a small variable voting pool. RESULTS: An unsupervised CRF model is proposed for efficient analysis of gene expression time series and is successfully applied to gene class discovery and class prediction. The proposed model treats each time series as a random field and assigns an optimal cluster label to each time series, so as to partition the time series into clusters without a priori knowledge about the number of clusters and the initial centroids. Another advantage of the proposed method is the relaxation of independence assumptions.

BACKGROUND: Tight clustering arose recently from a desire to obtain tighter and potentially more informative clusters in gene expression studies. Scattered genes with relatively loose correlations should be excluded from the clusters. However, in the literature there is little work dedicated to this area of research. On the other hand, there has been extensive use of maximum likelihood techniques for model parameter estimation. By contrast, the minimum distance estimator has been largely ignored. RESULTS: In this paper we show the inherent robustness of the minimum distance estimator that makes it a powerful tool for parameter estimation in model-based time-course clustering. To apply minimum distance estimation, a partial mixture model that can naturally incorporate replicate information and allow scattered genes is formulated. We provide experimental results of simulated data fitting, where the minimum distance estimator demonstrates superior performance to the maximum likelihood estimator. Both biological and statistical validations are conducted on a simulated dataset and two real gene expression datasets. Our proposed partial regression clustering algorithm scores top in Gene Ontology driven evaluation, in comparison with four other popular clustering algorithms. CONCLUSION: For the first time partial mixture model is successfully extended to time-course data analysis. The robustness of our partial regression clustering algorithm proves the suitability of the combination of both partial mixture model and minimum distance estimator in this field. We show that tight clustering not only is capable to generate more profound understanding of the dataset under study well in accordance to established biological knowledge, but also presents interesting new hypotheses during interpretation of clustering results. In particular, we provide biological evidences that scattered genes can be relevant and are interesting subjects for study, in contrast to prevailing opinion.

As medical image databases are connected together through PACS, those medical images may suffer from security breaches if not protected. If a medical image is illegally obtained or if its content is malevolently changed, the patient's privacy or health care to be provided will certainly be undermined. To solve this problem, this study proposes a steganographic method for mammograms, which hides patients' information in mammograms without changing their important details. In particular, the proposed method can extract the hidden information from stego-mammograms without the aid of the original images. This study also utilizes a watermarking technique, which masks the contents of the mammogram, for providing mammograms protection against illegal access and malevolent modifications. This watermark can be removed to reveal the masked mammogram when authorization for viewing is given.

In this paper, a novel semi-fragile authentication scheme based on block-mean quantization in the wavelet domain is proposed for image authentication. The scheme is composed of a structural digital signature process and a semi-fragile watermarking/ fingerprinting algorithm. In the signature process, the invariants are extracted as authentication codes from the quantization relationships and hierarchy information of discrete wavelet decomposition. The authentication codes consist of inter-block and intrablock codes in order to detect and localize tampering. The semi-fragile fingerprinting algorithm is intended to embed the signature via the use of block-mean quantization. The proposed scheme can effectively detect and localize the malicious modification while tolerating lossy compression.

In this paper we demonstrate the inherent robustness of minimum distance estimator that makes it a potentially powerful tool for parameter estimation in gene expression time course analysis. To apply minimum distance estimator to gene expression clustering, a partial mixture model that can naturally incorporate replicate information and allow scattered genes is formulated specially for tight clustering. Recently tight clustering was proposed as a response for obtaining tighter and thus more informative clusters in gene expression studies. We provide interesting results through data fitting when compared with maximum likelihood estimator using simulated data. The experiments on real gene expression data validated our proposed partial regression clustering algorithm. Our aim is to provide interpretations, discussions and examples that serve as resources for future research.

Breast cancer is one of the most common diseases among women. Content-based mammogram retrieval has been proposed to aid various medical procedures. To develop a content-based mammogram retrieval system, textural feature extraction is one of the crucial requirements. This study proposes a Gabor filtering method for the extraction of textural features, which firstly performs Gabor filtering on the underlying image, applies the physical properties of a probability wave to probability transformation and then computes features to describe the textural pattern of the mammogram. This study also proposes an adaptive strategy for feature selection, filter selection and feature weighting, which utilizes a user's relevance feedback to reduce the redundancy in the representation and incorporates the user's information needs in image retrieval. Experimental results show that hypothesis tests can effectively find discriminated features and this retrieval system can improve its performance through just a few rounds of relevance feedback.

We propose in this work a method for detecting faces in colour images with complex backgrounds. The approach starts with the transformation of the image pixels from the RGB colour space to the chrominance space (YCbCr). Secondly, a Gaussian model is fitted on the transformed image in order to calculate the likelihood of skin for each pixel and to create a likelihood image. Thirdly, by thresholding the likelihood image, skin pixels are segmented to form a binary skin map, which contains the candidate face regions. Finally, a verification process is carried out to determine whether these candidate face regions are real faces or not.

In this work, we propose a steganographic scheme for embedding secret message in VQ-compressed images, which aims to achieve high embedding capacity while keeping distortion low. To achieve these objectives, we develop a progressive exponential clustering (PEC) algorithm for partitioning the VQ codebook into a set of clusters. To embed the secret message based on the clustering, we substitute the indexes of the codewords with others within the same cluster according to the size of the cluster and the secret message bits.

In this paper, we propose a CBIR approach to retrieve building images. First, we use the Canny edge detector to extract edge information from the images. Secondly, the Hough transform is applied to the edge map in order to reveal the linear edge distribution in the Hough transform domain. Thirdly, by using a Band-wise matching (BWM) algorithm, we partition the Hough transform domain into a number of bands and calculate the centroid of the Hough peaks in each band. By carrying out the same aforementioned procedures on a query image and the images in the database, we measure the similarity between the centroids of the query image and the images in the database. Finally, based on the similarity measures, the CBIR system ranks the images in the database and retrieves a specified number of images with the highest rankings.

Content-based image retrieval (CBIR) has been proposed to address the problem of image retrieval from medical image databases. Relevance feedback, explaining the user's query concept, can be used to bridge the semantic gap and improve the performance of CBIR systems. This paper proposes a learning method for relevance feedback, which utilizes probabilistic model to generalize the 2-class problem and provide an estimate of probability of class membership. To build the probabilistic model, support vector machine (SVM) is applied to classify the mammograms, and then scale them to the probability of class membership. Experimental results show that the proposed learning method can effectively improve the average precision rate from 40% to 62% through five iterations of relevance feedback rounds

In this paper we propose an efficient data embedding algorithm for palette-based images by quantising similar colours in the palette. Unlike previous algorithms, this method's rate-distortion behaviour is independent from the embedding message, so that given the length of the embedding message, the distortion on the host image can be determined before embedding. Experiments show that our method outperforms previous methods both in terms of PSNR and visual quality.

This work approaches the texture segmentation problem by incorporating genetic algorithm and k-mean clustering method within a multiresolution structure. First, a quad-tree structure is constructed and the input image is partition into blocks at different resolution levels. Texture features are then extracted from each block. Based on the texture features, a hybrid genetic algorithm is employed to perform the segmentation. The crossover operator of traditional genetic algorithm is replaced with k-means clustering method while the mutate and select operators are adopted. In the final step, the boundaries and the segmentation result of the current resolution level are propagated down to the next level to act as contextual constraints and the initial configuration of the next level, respectively.

In this paper, a class of Random Field model, defined on a multiresolution array is used in the segmentation of gray level and textured images. The novel feature of one form of the model is that it is able to segment images containing unknown numbers of regions, where there may be significant variation of properties within each region. The estimation algorithms used are stochastic, but because of the multiresolution representation, are fast computationally, requiring only a few iterations per pixel to converge to accurate results, with error rates of 1-2 percent across a range of image structures and textures. The addition of a simple boundary process gives accurate results even at low resolutions, and consequently at very low computational cost.

An efficient fragile watermarking scheme intended for image authentication and integrity verification is proposed in this work. To watermark the underlying image, the gray scale of each pixel is adjusted by an imperceptible quantity according to the consistency between a key-dependent binary watermark bit and the parity of a bit stream converted from the gray scales of a secrete neighborhood of the pixel. To counter “collage” and “look-up table inferring” attacks, the scanning / watermarking order of the pixels follows a zig-zag path and the secrete neighborhood of a pixel is formed by picking previously watermarked pixels before the current pixel on the scanning path.

A technique using the inherent feature map of the underlying image as the watermark is proposed in this work. First, the binary feature map is extracted as watermark and partitioned into blocks. Secondly, neighboring feature map blocks are blended and encrypted for insertion. On the receiver side, the feature map from the received image is extracted again and compared against the recovered watermark to verify the integrity and authenticity. In addition the capability of detecting geometric transformation, removal of original objects and addition of foreign objects, the proposed scheme is also capable of localizing tampering and detecting cropping without a priori knowledge about the image.