Online review platforms have drastically reshaped how we interact, make purchasing decisions, and engage with digital content. However, the rise of deceptive content, privacy breaches, and misinformation has undermined online content credibility, impacting the trustworthiness of the information shared. To address these issues, we propose a robust framework for automated trust assessment of online reviews, focusing on identifying deceptive online content. The core of our approach is the trust vector, a novel feature representation that captures key user engagement factors influencing content trustworthiness. By applying the Weighted Trust Scoring Method (WTSM), we calculate a weighted trust score that strengthens the model’s interpretability and effectiveness in trust evaluation. The proposed model is evaluated on two benchmark datasets-the Deceptive Opinion Corpus Dataset and the Yelp Review Dataset. The framework achieves classification accuracies of $85 \%$ and $87 \%$, respectively, demonstrating its effectiveness in distinguishing deceptive from truthful content.

In an era of increasing digital privacy risks, images shared online can inadvertently reveal sensitive location data through identifiable elements such as logos, road signs, and text. These disclosures enable unauthorized tracking and data mining, raising serious privacy concerns. This paper proposes a novel framework that combines object detection and generative inpainting for privacy preserving image reconstruction. A Mask R-CNN model is trained on three diverse datasets to detect and segment location identifiable elements accurately. The detected regions are then de-identified using a proposed Denoising Diffusion Probabilistic Model (DDPM)-based inpainting method, which preserves scene integrity by ensuring geometric consistency and natural lighting. Unlike traditional inpainting methods, the proposed framework dynamically refines image reconstructions through controlled denoising, achieving high realism. The effectiveness of the method is evaluated using standard image quality metrics, including PSNR, SSIM, and FID, alongside subjective visual assessments. Experimental results show that the proposed approach outperforms baseline models such as CNNs and GANs, offering a robust solution for privacy preserving image reconstruction.

User identification through aesthetic preferences has gained attention as a promising direction in social behavioral biometrics, offering a non-invasive and privacy-conscious alternative to traditional physiological identifiers. Unlike still images or single-modal inputs, video-based aesthetic preferences provide richer, temporally-aware insights into user behavior, enabling a deeper understanding of personal taste and style. Despite this potential, existing approaches often treat preference items independently and fail to capture the interrelationships within a user’s preference set. To address these limitations, this paper introduces Preference-Aware Set Encoding with Contrastive Personalization (PASE), a novel deep learning framework designed to model user identity based on structured preferred video sets. The proposed method integrates a Cross-Video Attention Encoder to learn co-preference patterns across videos, a User-Anchor Contrastive Loss to align personalized embeddings, and a Cross-Set Mixup Regularization technique to improve generalization by simulating diverse preference scenarios. Evaluation on a curated aesthetic dataset demonstrates that PASE achieves 98.38% identification accuracy, outperforming unimodal and multi-modal baselines. These findings highlight the unique advantages of leveraging video-based aesthetic information for biometric identification, particularly in applications demanding both accuracy and user-friendly privacy safeguards.

We propose the Style-Preserving Generator (SPG) to generate synthetic license plate data to train License Plate Recognition (LPR) models, and compare the performance with the same models trained on real-world data. The proposed SPG can edit the characters on real-world license plates while maintaining their original styles, allowing synthetic license plate data to be generated with user-specified characters. We can therefore synthesize license plates with desired characters to effectively alleviate the data attribute imbalance and privacy issues associated with real-world license plates. To the best of our knowledge, this work is the first study to present the making of synthetic LP data by proposing a novel text-editing approach tailor-made for LP data, that is the proposed SPG. The SPG consists of a transformer, a source encoder, a source style encoder, a character mask decoder, a target generator, and a target discriminator. Given a source license plate image and a specified text as input, these components collaborate to compute the self- and cross-attention embeddings, predict character masks, and generate a synthetic license plate in the source style but with source characters replaced by the specified characters. We adopt a two-phase training scheme. Phase 1 training uses synthetic data only, but Phase 2 training uses synthetic and real-life data. To showcase the effectiveness of the SPG, we introduce a new benchmark dataset, the LP-2025 (License Plate 2025), which alleviates the limitations of existing datasets and presents new challenges for license plate recognition and generative models. We validate SPG performance on the LP-2025 dataset and other benchmark datasets and compare it against state-of-the-art text-editing approaches.

Soft biometric de-identification is an emerging field in biometrics, offering a balance between privacy protection and recognition accuracy. In this work, we present a novel identity-preserving soft biometric obfuscation method that uses the latent feature space of a trained generator and employs deep neural networks. The proposed method aims at preserving the identity of individuals while de-identifying their soft biometric attributes. Specifically, a novel feature space transformation network is designed to preserve identity while modifying facial attributes while minimizing the disclosure of identity. The proposed feature transformation network is the first of its kind developed specifically for controllable adaptive de-identification. Furthermore, we implemented an identity preservation mechanism, utilizing the FaceNet architecture to compute embedding vectors for both the original and deidentified images. Through extensive validation on benchmark datasets such as VGGFace2 and CelebA, we have demonstrated the effectiveness and robustness of our method.

Transformer architectures emerged as frontrunning approaches for computer vision classification tasks. Although transformers are faster than recurrent networks, they can become time and memory-inefficient. This paper proposes a novel filtering strategy based on Laguerre Voronoi diagram-based feature similarity measure. The approach allows to choose the most influential tokens and thus to reduce redundant information processing and extra computational costs. The filtering strategy can significantly reduce computation and memory overhead of vision transformers, depending on the number of patches to process. By comparing the proposed strategy with existing token filtering strategies, we report that the proposed architecture performs better than the comparators while reducing the computational overhead.

Responsible, ethical and trustworthy decision making powered by the new generation of artificial intelligence (AI) and deep learning (DL) recently emerged as one of the key societal challenges. The tutorial discusses key challenges, lists major applications, presents mitigation strategies and provides insights on the future developments in this emerging research domain.

Abstract Human aesthetics play a significant role in video game development, emotional‐aware robot design, online recommender systems, digital human, and other domains of research focusing on human‐computer interactions. Social network user recognition based on aesthetic preferences is an emerging research domain. In this paper, a novel deep learning architecture is proposed for multi‐modal audio‐visual person identification that combines audio and visual aesthetic features. A pre‐trained ResNet architecture is utilized to extract high‐level features from a set of user‐preferred audio and image samples. A novel deep learning‐based fusion technique called residual‐aided intermediate fusion (RAIF) is introduced in order to effectively merge the audio and visual features. The proposed RAIF method achieved an accuracy of 98% and a loss of 0.01 on a proprietary multi‐modal dataset, indicating its effectiveness in fusing audio and visual information.

Online social media (OSM) has become an integral part of an individual’s daily life. The extensive computational power and decision-making ability of artificial intelligence (AI) and the proliferation of user-generated data on OSM have made the opinion one of the key emerging research areas. However, the ease of accessing, manipulating, and mining such user-generated data raises concerns about privacy and security, data and algorithmic biases and fairness. Nevertheless, their personal and societal implications are barely addressed. In this paper, we discuss the limitations, fairness, and biases introduced in data mining and the AI model development. Moreover, we describe the possible implications of using AI systems on users’ privacy and address future research directions to mitigate potential biases.

Recognizing author identity from digital footprints without having a large corpus of documents from an individual is of keen interest to security researchers and government agencies. Users reveal aspects of their personality via the content they share with their social media followers and through the patterns in their interactions on online networking platforms. This study examines the potency of emerging natural language processing (NLP) methods in analyzing social network activity. A linguostylistic personality traits assessment (LPTA) system is developed to estimate Twitter users’ personality traits based on their tweets using the Myers-Briggs-type indicator (MBTI) and big-five personality scales. A novel input representation mechanism is proposed to process tweets by converting them into real-valued vectors using frequency, co-occurrence, and context (FCC) measures. Other prevalent text representation schemes, such as one-hot encoding, count-based vectorization, and pretrained language model representations are used as comparators. A genetic algorithm (GA) approach is proposed to reduce the feature set and increase the efficacy of the features extracted. The developed system outperforms the state-of-the-art research by reliably estimating the user’s latent personality traits while using 50 or fewer tweets per user.

Aesthetic preference can be described as one's taste or fondness for a particular subject. This information has become ubiquitous as online communities and social media have grown increasingly integrated with daily life. The domain of social-behavioral biometrics analyzes the interactions, relations, and communications of individuals rather than traditional physical traits. Recent research has demonstrated that a person's visual aesthetic preferences possess discriminatory value for person identification. This paper introduces the first audio and visual multi-modal aesthetic identification system that utilizes both user-liked images and songs for an accurate identity prediction with score-level fusion. The developed multimodal system achieves an accuracy of 99.4% on the proprietary audio-visual dataset, outperforming unimodal systems.

The problem of user recognition in online social networks is driven by the need for higher security. Previous recognition systems have extensively employed content-based features and temporal patterns to identify and represent distinctive characteristics within user profiles. This work reveals that semantic textual analysis and a graph representation of the user’s social network can be utilized to develop a user identification system. A graph transformer network architecture is proposed for the closed-set node identification task, leveraging the weighted social network graph as input. Users retweeting, mentioning, or replying to a target user’s tweet are considered neighbors in the social network graph and connected to the target user. The proposed user identification system outperforms all state-of-the-art systems. Moreover, we validate its performance on three publicly available datasets.

Face normalization can be a crucial step when handling generic face recognition. We propose the Pose and Illumination Normalization (PIN) framework with contrast data learning for face normalization. The PIN framework is designed to learn the transformation from a source set to a target set. The source set and the target set compose a contrastive data set for learning. The source set contains faces collected in the wild and thus covers a wide range of variation across illumination, pose, expression and other variables. The target set contains face images taken under controlled conditions and all faces are in frontal pose and balanced in illumination. The PIN framework is composed of an encoder, a decoder and two discriminators. The encoder is made of a state-of-the-art face recognition network and acts as a facial feature extractor, which is not updated during training. The decoder is trained on both the source and target sets, and aims to learn the transformation from the source set to the target set; and therefore, it can transform an arbitrary face into a illumination and pose normalized face. The discriminators are trained to ensure the photo-realistic quality of the normalized face images generated by the decoder. The loss functions employed in the decoder and discriminators are appropriately designed and weighted for yielding better normalization outcomes and recognition performance. We verify the performance of the propose framework on several benchmark databases, and compare with state-of-the-art approaches.

A typical biometric system aims to recognize individuals based on their unique physiological or behavioral traits. Online Social Networking (OSN) platforms have become an integral part of the daily life of individuals, where they leave a recognizable trail of behavioral information. Social Behavioral Biometric (SBB), being an emerging trend, focuses on such trails to distinguish between individuals. This research investigates the impact of users' writing profiles on OSN to conclude whether such profiles contribute to SBB. The distinctiveness of the SBB features that are extracted from the social behavioral data of Twitter is studied. A person identification system that relies on the writing profiles of OSN users is proposed. The developed system is cross-validated on a social interaction database of 241 Twitter users. The rank-1 identification rate from users' writing profiles is 91.70% and the rank-8 identification rate is 99%. Furthermore, the experimental results establish that the users' writing profiles have the highest impact over other social biometric features.

It is recognized that system trustworthiness is a hyperstructure embodied by the structural, behavioral, and system dimensions with a set of coherent attributes. We explore a theoretical framework of tripartite trustworthiness that can be applied to real-world autonomous systems. We present a formal study of the essences and mathematical models of system trustworthiness and their quantitative measurements in the contexts of autonomous and mission-critical intelligent systems where humans and machines interact in a hybrid environment.

Rapid growth of social networks has provided an extraordinary medium to share a large volume of photographs online. This calls for designing efficient face recognition techniques that are applicable to images with low resolutions and arbitrary poses. This paper proposes a new pose invariant face recognition method for low resolution images using only a single training sample. A 3D model, reconstructed using Generic Elastic Model (3D GEM) from a frontal view training sample, is used to generate a set of nonfrontal gallery face images. The face region of the nonfrontal query sample is then extracted using the same landmark detection technique as in the 3D GEM algorithm. Afterwards, a novel texture representation technique called Local Comparative Decimal Pattern (LCDP) is proposed to extract features from each of the training and query samples. A set of experimental results on the ORL, Georgia Tech (GT), and LFW face databases demonstrates the efficiency of the proposed method compared to other state-of-the-art approaches.

Personality traits have been shown to have strong influences on important aspects of life such as success in the workplace, political temperament, and general emotional stability. Computer-based personality assessments using information from social networking platforms have shown to be more accurate than judgments made by people close to the subject. This paper presents a personality traits classification system that incorporates language-based features, based on count-based vectorization (TF-IDF) and the GloVe word embedding technique, with an ensemble prediction system consisting of gradient-boosted decision trees and an SVM classifier. This combination allows to reliably estimate certain personality traits using only the latest 50 tweets from a user's profile. The performance of the proposed system is validated on a large, publicly available dataset and compares favourably with other state-of-the-art methods.

The last decade has witnessed an increase in online human interactions, covering all aspects of personal and professional activities. Identification of people based on their behavior rather than physical traits is a growing industry, spanning diverse spheres such as online education, e-commerce and cyber security. One prominent behavior is the expression of opinions, commonly as a reaction to images posted online. Visual aesthetic is a soft, behavioral biometric that refers to a person's sense of fondness to a certain image. Identifying individuals using their visual aesthetics as discriminatory features is an emerging domain of research. This paper introduces a new method for aesthetic feature dimensionality reduction using gene expression programming. The advantage of this method is that the resulting system is capable of using a tree-based genetic approach for feature recombination. Reducing feature dimensionality improves classifier accuracy, reduces computation runtime, and minimizes required storage. The results obtained on a dataset of 200 Flickr users evaluating 40000 images demonstrates a 94% accuracy of identity recognition based solely on users' aesthetic preferences. This outperforms the best-known method by 13.5%.

Cancelable or Revocable biometrics is a recent trend to safeguard a biometric system from a variety of attacks. In this paper, we propose a cancelable system in which iris features are extracted through deep learning and then converted into a cancelable biometric template through random projection method. We then adopt another machine learning algorithm - Support Vector Machine for optimal biometric authentication after performing comparitive analysis over 3 alternative classifiers. The proposed system provides better template security and improves identification accuracy. As per our knowledge, this combination of deep learning and random projection technique has been employed for the first time. The paper presents an extensive validation of the proposed methodology on two multi-instance iris databases.

Human brain has an ability to perform a massive processing of auxiliary information such as visual cues, cognitive and social interactions, contextual and spatio-temporal data. Similarly to a human brain, social behavioral cues can aid the reliable decision-making of a biometric security system. Being an integral part of human behavior, social interactions are likely to possess unique behavioral patterns. This state-of-the-art review paper discusses an emerging person recognition approach based on the in-depth analysis of individuals' social behavior in order to enhance the performance of a traditional biometric system. The social behavioral information can be mined from their offline or online interactions, and can be identified as a set of Social Behavioral Biometric (SBB) features. These features could be used on their own or further combined with other behavioral and physiological patters, and classification can be enhanced by the use of machine learning approaches. An overview of open problems and challenges as well as applications of studying social behavior in various domains concludes this paper.

Recognition from body movement is a challenging domain of research that lies at an intersection of machine learning, biometric security and cognitive functions domain. It can be highly beneficial for expert systems, lie detectors, border control, medical emergencies, as well as search and rescue operations. This paper describes a first prototype of a real-time system capable of recognizing four gestures that correlate to human emotions based on the arm movements. Features extracted from the 3D skeleton using Kinect v2 sensor are classified using an SVM method. The system is tested in real-time on a Kinect database with the embedded system using an optimized algorithm for skeleton extraction in real-time.

Social interactions via Online Social Network (OSN) can provide a gamut of information about users that have been recently studied as behavioral patterns for person recognition. Similar to social interactions, the temporal information of persons in OSN is likely to exhibit behavioral characteristic and habitual pattern. This paper presents the first empirical study to answer a question whether temporal information obtained via OSN may contain sufficient behavioral biometric properties. In this paper, we present a methodology to identify a set of idiosyncratic temporal features and develop a system based on those unique features for identity verification. To the best of our knowledge, this is the first study on identity verification based on solely temporal profile obtained from an online social network. Experiments demonstrate that the proposed unique temporal profile in OSN can be utilized for users' identity verification, as it obtained low EER of 12% and high AUC of 95.2% in a closed-set test scenario. Potential applications of the proposed temporal profile include identity verification, anomaly and fraud detection, identity theft, continuous authentication, human behavior analysis, and so on.

Brain-Inspired Systems (BIS) are an emerging field of brain and intelligence sciences that studies natural intelligence models of AI and cognitive systems in one direction, and the formal models of the brain simulated by computational intelligence in another direction. A typical BIS is the cognitive robots that mimic and implement the brain through all cognitive levels. BIS provides insights for brain-machine interfaces (BMI), which may lead to novel man-machine interactions and hybrid intelligent systems. BIS may also advance classic computers from dada processors to the next generation of knowledge processors mimicking the brain. BIS will underpin a wide range of engineering paradigms such as cognitive systems, cognitive computers, cognitive robots, machine learning systems, semantic comprehension systems, big data systems, unmanned systems, self-driving vehicles and hybrid man-machine systems.

The goal of a biometric recognition system is to make a human-like decisions on individual's identity by recognizing their physiological and/or behavioral traits. Nevertheless, the decision-making process by either a human or a biometric recognition system can be highly complicated due to low quality of data or an uncertain environment. Human brain has an advantage over computer system due to its ability to perform a massive parallel processing of auxiliary information, such as visual cues, cognitive and social interactions, contextual, and spatio-temporal data. Similarly to a human brain, social behavioral cues can aid the reliable decision-making of an automated biometric system. In this paper, a novel person recognition approach is presented, that relies on the knowledge of individuals' social behavior to enhance the performance of a traditional biometric system. The social behavioral information of individuals' has been mined from an online social network and fused with traditional face and ear biometrics. Experimental results on individual's and semi-real databases demonstrate significant performance gain in the proposed method over traditional biometric system.

Automatic license plate detection (ALPD) is one of the widespread research topics in the area of intelligent transportation systems. A challenging issue that affects ALPD performance is the complex image background, where possibility of misclassification of non-license plate (non-LP) objects as a license plate (LP) object is high. One of the ways to resolve the issue is to use an efficient filter to correctly classify the license plate and non-license plate objects. In this paper, we propose an efficient general technique for the classification of LP and non-LP images based on Histogram Oriented Gradient (HOG) features, and mixture of experts model (binary classifiers). To maximize the classification performance, Genetic Algorithm (GA) is applied to find the best feature subset and adjust the weights of the mixture model. Performance of the proposed method is evaluated on a new database of 2360 LP and non-LP images created by us. Experimental results achieve image classification with a high accuracy of 85.16%. The filter is also tested by plugging it into a recent ALPD system which improves the detection performance by 6.7%.

Quality of facial images has a great impact on the accuracy of the automated face recognition system. The intraclass variations introduced by varied facial quality due to variation in illumination conditions may degrade the performance of a face recognition system significantly. In this paper, we proposed an adaptive discrete wavelet transform (DWT) based face recognition approach which will normalize the illumination distortion using regional contrast limited adaptive histogram equalization (CLAHE) and discrete cosine transform (DCT) normalization based on the illumination quality. The DWT based approach is used to extract the low and high frequency facial features at different scales. In the proposed method, a weighted fusion of the low and high frequency subbands is computed to improve the identification accuracy under varying lighting conditions. The selection of fusion parameters is made using fuzzy membership functions. The performance of the proposed method was validated on the Extended Yale Database B. Experimental results depict that the proposed method outperforms some well known face recognition approaches.

Authorship recognition from micro-blogs such as Twitter is a challenging task due to limitation of text length to 140 characters. However, identification of micro-blog authors is crucial in many cyber-crime investigations as well as in forensic applications. So far, traditional linguistic profiles such as Bag-Of-Words (BOW) and style-based markers have been investigated for identification of micro-blog authorship. The social interactive data in micro-blogs remained understudied for this purpose. In this paper, we examined authorship recognition based on the social interactions of users in Twitter and present a comparative analysis with BOW and style-based features. We obtained 97% recognition rate on a database of 70 Twitter users, which validates the superiority of using social interactive data compared to traditional linguistic profiles.

Social interactions are integral part of human behavior. Although social interactions are likely to possess unique behavioral patterns, their significance for automated user recognition has been noted in the scientific community only recently. This paper demonstrated that it is possible to generate a set of unique features, called social behavioral (SB) features, from the social interactions of individuals' via an online social network (OSN). Specifically, this research identified a set of SB features from the online social interactions of 241 Twitter users and proposed a framework to utilize these features for an automated user recognition. Extensive experimentation demonstrated high recognition performance as well as distinctiveness of the proposed SB features. The most striking finding was that only ten recent tweets are enough to recognize 58% of users in our database at rank-1. The rank-1 recognition rate dramatically increased to 93% when 60 tweets were used as a probe set. Experimental results also demonstrated the stability of the proposed SB feature set over time and ability to recognize both frequent and nonfrequent OSN users. This confirms that human social behavior expressed through an OSN can provide a unique insight into user behavior recognition.

This paper presents a new action recognition method that utilizes the 3D skeletal motion data captured using the Kinect depth sensor. We propose a robust view-invariant joint motion representation based on the spatio-temporal changes in relative angles among the different skeletal joint-triplets, namely the joint relative angle (JRA). A sequence of JRAs obtained for a particular joint-triplet intuitively represents the level of involvement of those joints in performing a specific action. Collection of all joint-triplet JRA sequences is then utilized to construct a spatial holistic description of action-specific motion patterns, namely the 2D joint-triplet motion image. The proposed method exploits a local texture analysis method, the local binary pattern (LBP), to highlight micro-level texture details in the motion images. This process isolates prototypical features for different actions. LBP histogram features are then projected into a discriminant Fisher-space, resulting in more compact and disjoint feature clusters representing individual actions. The performance of the proposed method is evaluated using two publicly available Kinect action databases. Extensive experiments show advantage of the proposed joint-triplet motion image and LBP-based action recognition approach over existing methods.

Cybersecurity is an important and challenging issue faced by governments, financial institutions and ordinary citizens alike. Secure identification is needed for accessing confidential government information, online bank transaction, person's social network (Facebook, Twitter, Linkedin). Brain signal electroencephalogram (EEG) can play a vital role in ensuring security as it is non-vulnerable and hard to steal. In this article, we develop an EEG based biometric security system. The purpose of this work is to find the best band or the best bands combination of overt mental stimuli of brain EEG signal to identify a person. The Discrete Wavelet Transform (DWT) is used to extract different significant features which separate Alpha, Beta and Theta band of frequencies of the EEG signal. Extracted EEG features of different bands and their combinations such as alpha-beta, alpha-theta, theta-beta, alpha-beta-theta are classified using an artificial neural network (ANN) trained with the back propagation (BP) algorithm. The classification rate shows that Alpha band (84.4%) has higher mapping precision and better convergence rate than the other bands, beta (80%), theta (78.1%) and bands combination as alpha-beta (64.1%), alpha-theta (65.6%), beta-theta (58.8%), alpha-beta-theta (56.9%). Another classifier K nearest neighbor (KNN) is used to verify this result. The classification result of this KNN classifier also shows that alpha band (50%) has higher convergence rate than other bands, beta (40%) and theta (40%). The results of this study are expected to be helpful for future research of overt mental stimuli brain signal based biometric approaches.

Faces captured in a real-world scenario may suffer from large variations in shape and occlusions due to difference in illumination, variation in pose and orientation of a facial image. Automated face recognition or security reinforcement by surveillance techniques would be useless if the faces are occluded. Therefore, face occlusion detection has become very important not only for effective face recognition but also to prevent security threats. In this paper, for the very first time an occlusion detection method is proposed based on the depth information provided by Kinect RGB-D cameras. Uniform Local Binary Pattern (LBP) is used to effectively extract the features from the depth images and SVM binary classifier is then applied to identify the front face and the occluded face. For localizing occluded regions in the face image, a threshold based approach is proposed to identify the areas close to the camera. In the depth images, an object close to the camera has a higher pixel intensity than the object further from the camera. Thus, we assume that occluded regions have lower distance from the camera, i.e. higher intensity values. Based on this hypothesis, we extract the connected component with highest energy values as the potential occluded region from the depth image. The boundary of the detected occluded region is then corrected using the reference front face image. The occlusion detection and localization method have been evaluated on EUROKOM Kinect face database containing different types of occluded and unoccluded faces with neutral expressions. Experimental results show that the proposed method provides an average detection rate of 98.50% for front and occluded face images. We have also compared our proposed method with existing methods that use faces acquired using 3D scanners for occlusion detection.

This paper presents a model-based action recognition system that utilizes the Kinect 3D skeleton to construct adaptive spatio-temporal motion representations. The proposed method utilizes two features, namely the joint relative distance (JRD) and joint relative angle (JRA) to encode the spatio-temporal motion patterns of different skeletal joints. To evaluate the relevance of a particular joint-pair in representing an action class, we introduce a flatness measure that quantifies the level of engagement of the corresponding joint-pair in performing the action. The flatness measures computed for all skeletal joint-pairs are accumulated to construct a joint-pair relevance (JPR) matrix, which facilitates adaptive pooling of the most relevant spatio-temporal features to construct the final motion description for individual action classes. In addition, we propose a score level fusion of JRD and JRA features with a weighted dynamic time warping (DTW)-based matching scheme to effectively boost the overall recognition performance. In our experiments, the proposed method achieves better recognition performance than well-known existing methods.

Ear biometrics are often partially or fully occluded by hair, earrings, headphones, hat/cap, scarf, and other obstacles. Occurrence of occlusion during identification stage may cause significant information loss, which deteriorates recognition performance. In this paper, we proposed a novel index-based rank fusion method for ear recognition that can utilize occlusion information adaptively during identification stage to decide on a person's identity. In the proposed method, feature sets are selected and weighted according to the proportion of occlusion during identification time. Our experimental results on wide variety of real as well as synthetically occluded ears demonstrate that the proposed adaptive feature selection and fusion method significantly improves the recognition performance of occluded ears.

This paper presents a simple, yet effective facial feature descriptor based on evolutionary synthesis of different local texture patterns. Unlike the traditional face descriptors that exploit visually-meaningful facial features, the proposed method adopts a genetic programming-based feature fusion approach that utilizes different local texture patterns and a set of linear and nonlinear operators in order to synthesize new features. The strength of this approach lies in fusing the advantages of different state-of-the-art local texture descriptors and thus, obtaining more robust composite features. Recognition performance of the proposed method is evaluated using the Cohn-Kanade (CK) and the Japanese female facial expression (JAFFE) database. In our experiments, facial features synthesized based on the proposed approach yield an improved recognition performance, as compared to some well-known face feature descriptors.

Automated recognition of facial expression has attracted significant attention in recent years due to its potential applicability in security and surveillance, human computer interaction, social robotics, and animation. This paper presents a new facial expression recognition method that utilizes bit plane specific local image description in a weighted score level fusion. The motivation is to utilize bit plane slicing to highlight the contribution of a particular bit plane made to the holistic facial appearance, which is then used in a weighted score level fusion in order to boost the recognition performance. A new local image descriptor is proposed specifically to extract local features from bit plane representations that utilizes Fisher linear discriminant to maximize the inter-class distance, while minimizing the intra-class variance. Two well-known facial expression databases, namely the Cohn-Kanade (CK) and the Japanese female facial expression (JAFFE) database have been used to evaluate the performance of the proposed method against existing facial appearance descriptors, such as local binary pattern (LBP), local ternary pattern (LTP), local directional pattern (LDP), and linear discriminant analysis (LDA). Experiments with a total of seven prototypic facial expressions show promising results for the proposed method, as compared with the other existing methods.

During this era of internet, crowd-sourcing is a very popular way of accommodating a large group of people contributing together to accomplish a goal. One of the most remarkable examples of such crowd sourced content is the Wikipedia, where millions of articles have been produced by volunteers from all over the world. Wikipedia allows anyone to edit articles without being authorized. Although creation of this huge repository of information is being possible because of the freedom of editing, it also attracts sock puppets and malicious users to cause ruthless destruction in Wikipedia contents. One way of dealing with such malevolent users is to predict the identity of ambiguous authors. However, authorship recognition in collaborative environment like Wikipedia is very challenging. In this paper, we propose a novel way of mapping ambiguous users identity to previously known users based on their editing profile. The proposed editing behavior based authorship recognition can be applied to decide on trusty and offensive authors, identity theft, shock puppetry, human behavior analysis, and so on. Our experimentation on a large database of Wikipedia demonstrate promising results of using editing behavior to recognize authors of collaborative writing.

In todays world, identity of human beings has expanded beyond the real world to the cyber world. Virtual identity of millions of users is present at various web-based Social Networking Sites (SNSs) such as Myspace, Facebook, and Twitter. Interactions through SNSs have become a part of our daily practices, which eventually leaves a big trail of behavioral pattern in virtual domain. In this paper, the authors examined the feasibility of person identification using such social network activities as behavioral biometrics. Experimentation includes extraction of a number of idiosyncratic features from SNSs and analysis of their performance as novel social behavioral biometric features.

Multimodal Biometric system is very proficient because of the advantageous aspects over unimodal biometric system. Feature fusion based multimodal system is one of the best in its genres because it only stores single template and decries the privacy and security threats as well as the system memory. However, biometric templates from traditional feature fusion for multi-biometric systems are vulnerable in terms of template protection, where it can only improve the performance. On the other hand, proposed cancelable fusion is a new type of feature fusion for multimodal biometric system that can achieve both improved performance for multimodality and cancelability at the same time. In other word, proposed cancelable fusion keeps all the characteristics of multimodal biometric systems and ensures the template security in addition so that hackers cannot use the multi-biometric template to break the authentication system even if the template is compromised.

A person can be identified from his physiological traits as well as from behavioral patterns. However, a person's behavior is not only confined to individual actions such as walking or typing style, speech or handwriting but also social interactions and communication. In other words, social communication is an indispensable part of our daily behavior. Therefore, a person's social connections, spatio-temporal information, style of interactions etc. Can be a good source of information to identify his social behavioral pattern. Based on this hypothesis, this paper introduces a novel kind of behavioral biometrics called Social Behavioral Biometrics (SBB) for the first time. The study includes identification of social behavioral biometric features from real and virtual domain and their prospective applications for the purpose of person authentication and verification.

Online Social Networking Sites (SNSs) are considered as one of the well-established mediums of mass communication in today's world. Similar to physical world humans tend to have unique pattern of social communication in virtual world. However, analysis of such web based communication patterns is rarely seen for person identification. Most of the existing biometric recognition systems use either individual physiological or behavioural traits. A framework for the analysis of the web-based social interaction data as biometric features is largely unexplored until now. In this paper, a framework to accumulate and analyze social communication based data from online SNSs is presented. Analysis of such features explores personal characteristics, knowledge, and communication patterns that can successfully be utilized as Social Behavioral Biometric features. Experimental results demonstrate that the proposed social behavioral biometric features are significantly useful for person authentication.

A novel Multi-Resolution Fusion (MRF) of Dual-Tree Complex Wavelet Transform (DTCWT) and Discrete Cosine Transform (DCT) is introduced in this paper. Shift invariant multi-scale feature set is obtained using 2D DTCWT. Subsequently, discriminant DCT coefficients are extracted to map the high dimensional features into low dimensional subspace. The resulting feature vector contains non-redundant discriminative information and is small in size. Therefore, the proposed face recognition technique exhibits computational efficiency, low storage requirement along with high recognition rate under varying shift conditions. It also provides robustness to expression and illumination change. The performance evaluation is accomplished on four standard face databases. Experimental results show significant performance improvement over existing well-established face recognition methods under varying conditions.

This paper presents for the first time decision fusion for multimodal biometric system using social network analysis (SNA). The main challenge in the design of biometric systems, at present, lies in unavailability of high-quality data to ensure consistently high recognition results. Resorting to multimodal biometric partially solves the problem, however, issues with dimensionality reduction, classifier selection, and aggregated decision making remain. The presented methodology successfully overcomes the problem through employing novel decision fusion using SNA. While several types of feature extractors can be used to reduce the dimension and identify significant features, we chose the Fisher Linear Discriminant Analysis as one of the most efficient methods. Social networks are constructed based on similarity and correlation of features among the classes. The final classification result is generated based on the two levels of decision fusion methods. At the first level, individual biometrics (face or ear or signature) are classified using matching score methodology. SNA is used to reinforce the confidence level of the classifier to reduce the error rate. In the second level, outcomes of classification based on individual biometrics are fused together to obtain the final decision.

In this paper, novel cancelable biometric template generation algorithm using Social Network Analysis is presented. Two sets of features are fused using Social Network. Proposed fusion technique is cancelable. Eigenvector centrality is used to generate final sets of features from the Virtual Social Network (VSN). The domain transformation of features using VSN confirms the cancelability in biometric template generation.

Cancelable biometric technique is one of the most effective methods of template protection. Essentially, the crucial security aspect of a biometric system is template protection. The concept behind the cancelable biometric or cancel ability is a transformation of a biometric data or extracted feature into an alternative form, which cannot be used by the imposter or intruder easily, and can be revoked if compromised. In this paper, we present a novel architecture for template generation in the context of situation awareness system in real and virtual application. We develop a novel cancelable biometric template generation algorithm utilizing random biometric fusion, random projection and selection. Proposed random cross-folding method generate cancelable biometric template from multiple bio-metric traits. We further validate the performance of the pro-posed algorithm using a virtual multimodal face and ear data-base.

In this paper, a novel face recognition technique based on discrete wavelet transform and Adaptive Local Binary Pattern (ALBP) with directional statistical features is proposed. The proposed technique consists of three stages: preprocessing, feature extraction and recognition. In preprocessing and feature extraction stages, wavelet decomposition is used to enhance the common features of the same subject of images and the ALBP is used to extract representative features from each facial image. Then, the mean and the standard deviation of the local absolute difference between each pixel and its neighbors are used within ALBP and the nearest neighbor classifier to improve the classification accuracy of the LBP. Experiments conducted on two virtual world avatar face image datasets show that our technique performs better than LBP, PCA, multi-scale Local Binary Pattern, ALBP and ALBP with directional statistical features (ALBPF) in terms of accuracy and the time required to classify each facial image to its subject.

Real-time locating systems (RTLS) have the potential to enhance healthcare systems through the live tracking of assets, patients and staff. This study evaluated a commercially available RTLS system deployed in a clinical setting, with three objectives: (1) assessment of the location accuracy of the technology in a clinical setting; (2) assessment of the value of asset tracking to staff; and (3) assessment of threshold monitoring applications developed for patient tracking and inventory control. Simulated daily activities were monitored by RTLS and compared with direct research team observations. Staff surveys and interviews concerning the system's effectiveness and accuracy were also conducted and analyzed. The study showed only modest location accuracy, and mixed reactions in staff interviews. These findings reveal that the technology needs to be refined further for better specific location accuracy before full-scale implementation can be recommended.

Dealing with high-dimensional data has an important role in a number of areas, including biometric recognition in both real world and emerging virtual reality applications. Acquiring a group of different biometrics with various characteristics and specifications results in a number of issues that should be addressed, while developing such multi-modal recognition system. In this paper, we propose a novel Multi-Modal Biometric System based on neural network paradigm which utilizes the ear and face features and has unique method to train different classifiers based on each feature set. The aggregation result depicts the final decision over the recognized identity. In order to train accurate set of classifiers, the subspace clustering method has been used to overcome the problem of high dimensionality of the feature space. The proposed system is based on a new methodology for shrinking down the finite search space of all possible subspaces by focusing on axis-parallel subspaces which is a novel approach in data clustering for biometric dataset. The experimental results over the FERET dataset show the superiority of the proposed method over several dimensionality reduction methods.

In this paper, we present a robust approach for face recognition from video sequences. An automatic face detectoris employed which uses modified skin color modeling to detect human skin regions from the video sequences. The presence or absence of face in each region is verified by means of height width proportion and a Neural Network based template matching scheme. The obtained face images are then projected onto a feature space, defined by Multilinear Principal Component Analysis (MPCA), to produce the biometric feature template. Recognition is performed by projecting anew image onto the feature spaces by the MPCA that generalizes not only the classical PCA solution but also a number of the so-called 2-D PCA algorithms and then classifying the face by comparing its position in the feature spaces with the positions of known individuals. The proposed method is applicable to security systems, secure human computer interaction, visual communication systems (secure video conferencing) and virtual world environments.

Art metrics, a field of study that identifies, classifies and authenticates virtual reality avatars and intelligent software agents, has been proposed as a tool for fighting crimes taking place in virtual reality communities and in multiplayer game worlds. Forensic investigators are interested in developing tools for accurate and automated tracking and recognition of avatar faces. In this paper, we evaluate state of the art academic and commercial algorithms developed for human face recognition in the new domain of avatar recognition. While the obtained results are encouraging, ranging from 53.57% to 79.9% on different systems, the paper clearly demonstrated that there is room for improvement and presents avatar face recognition as an open problem to the pattern recognition and biometric communities.

This issue features a special issue on ‘Computational Science and Its Applications’. Computational Science is the main pillar of most of the present research, industrial and commercial activities and plays a unique role in exploiting ICT innovative technologies. Owing to the latest development and the availability of high-performance computing, including parallel computing, grid computing, and cloud computing, there is a critical need to employ efficient and effective computational methods and algorithms in various applications, including computational biology, computational geometry, computational physics, computation chemistry, computational finance, graphics and visualization, scientific data management, data mining, etc. This issue features selected papers from the International Conference on Computational Science and Its Applications (ICCSA2009) held in 29 June–1 July, 2009, Kyung Hee University, Suwon, South Korea. In addition to extended papers from ICCSA2009, a special issue CFP has been distributed to a wider community through various mailing lists. Finally, we selected five papers to be included in this issue. The first paper discusses data and knowledge grids. It basically combines grid computing and real-time service management and execution paradigms, which makes use of the service-oriented architecture and paradigm suited for the grid platform. The second focuses on grid and P2P systems, especially in the context of sharing paradigm. It discusses middleware and libraries for grid and P2P systems. The third paper also focuses on P2P, whereby it describes scalable group communication protocols. The fourth paper focuses on road network query processing, which can be adopted in a mobile environment. It particularly concentrates on range search in a continuous mobile dynamic. Finally, the fifth paper introduces context-aware semantic network similarity model, using an ontological approach. As general co-chairs and program co-chairs of ICCSA2009, as well as the guest editors of the special issue on Computational Science in the Concurrency and Computation journal, we would like to congratulate the authors whose papers appeared in this special issue. We would also like to thank the PC members of ICCSA2009 who conducted the initial reviewing process for the conference and external reviewers who conducted further reviews of extended papers submitted to this special issue.

Geographic information systems (GISs) are widely used for representation, management, and analysis of spatial data in many disciplines. In particular, geoscientists increasingly use these tools for data integration and management purposes in many environmental applications, ranging from water resources management to the study of global warming. Beyond these capabilities, geoscientists need to model and simulate three-dimensional (3D) dynamic fields and readily integrate those results with other relevant spatial information in order to have a better understanding of the environmental problems. However, GISs are very limited for the modeling and simulation of spatial fields, which are mostly 3D and dynamic. These limitations are mainly related to the existing GIS spatial data structures that are static and limited to 2D space. In order to overcome these limitations, we develop and implement a new kinetic 3D spatial data structure based on Delaunay tetrahedralization and a 3D Voronoi diagram to support a 3D dynamic field simulation within GISs. In this article, we describe in detail the different steps from discretization of a 3D continuous field to its numerical integration, based on an event-driven method. For validation of the proposed spatial data structure itself and its potential for the simulation of a dynamic field, two case studies are presented in the article. According to our observations, during the simulation process, the data structure is maintained and the 3D spatial information is managed adequately. Furthermore, the results obtained from both experiments are very satisfactory and are comparable with the results obtained from other existing methods for the simulation of the same dynamic field. To conclude, we discuss the current challenges related to the development of the 3D kinetic data structure itself and its adaptation to 3D dynamic field simulation and suggest some solutions for its improvement.

Domestic and industrial robots, intelligent software agents, virtual world avatars and other artificial entities are quickly becoming a part of our everyday life. Just like it is necessary to accurately authenticate identity of human beings, it is becoming essential to be able to determine identities of non-biological agents. In this paper, we present the current state of the art in virtual reality security, focusing specifically on emerging methodologies for avatar authentication. We also outline future directions and potential applications for this high impact research field.

In this paper, an automatic rotation invariant multiview face detection method, which utilizes modified Skin Color Model (SCM), is presented. First, Gaussian Mixture Model (GMM) and Support Vector Machine (SVM) based hybrid models are used to classify human skin regions from color images. The novelty of the adaptive hybrid model is its ability to predict the chromatic skin color band for individual images based on calibration differences of camera and luminance condition of environment. Classified skin regions are then converted to gray scale image with a threshold based on the predicted chromatic skin color bands, which further enhances detection performance. Next, Principle Component Analysis (PCA) is applied to gray segmented regions. Face detection is carried out based on the PCA-based extracted features, along with selected features, using support vector regression. The output of this procedure is used to report the final result of face detection. The proposed method is also beneficial for the rotation invariant face recognition problem.

One of the most frequent queries in spatial and mobile databases is range search, which is originated from the construction of R-tree that limits the spatial database application to Euclidean distance. Nowadays, Geographic Information System (GIS) demands the applications to be practicable for factual distance, normally identified as network distance. Even though some algorithms are engaged in this area, network distance range search is still a time consuming and storage space occupation task. In this paper, we propose a novel approach which is based on Network Voronoi Diagram that is diffusely used in geometrical analysis. We are looking into how to improve the performance of range search query processing using Network Voronoi Diagram.

In many real-world applications, unimodal biometric systems often face significant limitations due to sensitivity to noise, intraclass variability, data quality, nonuniversality, and other factors. Attempting to improve the performance of individual matchers in such situations may not prove to be highly effective. Multibiometric systems seek to alleviate some of these problems by providing multiple pieces of evidence of the same identity. These systems help achieve an increase in performance that may not be possible using a single-biometric indicator. This paper presents an effective fusion scheme that combines information presented by multiple domain experts based on the rank-level fusion integration method. The developed multimodal biometric system possesses a number of unique qualities, starting from utilizing principal component analysis and Fisher's linear discriminant methods for individual matchers (face, ear, and signature) identity authentication and utilizing the novel rank-level fusion method in order to consolidate the results obtained from different biometric matchers. The ranks of individual matchers are combined using the highest rank, Borda count, and logistic regression approaches. The results indicate that fusion of individual modalities can improve the overall performance of the biometric system, even in the presence of low quality data. Insights on multibiometric design using rank-level fusion and its performance on a variety of biometric databases are discussed in the concluding section.

Facial expression modeling has been a popular topic in biometrics for many years. One of the emerging recent trends is capturing subtle details such as wrinkles, creases and minor imperfections that are highly important for biometric modeling as well as matching. In this paper, we suggest a novel approach to the problem of expression modeling and morphing based on a geometry-based paradigm. In 2D image space, a distance-based morphing system is utilized to create a line drawing style facial animation from two input images representing frontal and profile views of the face. Aging wrinkles and expression lines are extracted and mapped back to the synthesized facial NPR (nonphotorealistic) sketches. In 3D object space, we present a metamorphosis system that combines the traditional free-form deformation (FFD) model with data interpolation techniques based on the proximity preserving Voronoi diagram. With feature points selected from two images of the target face, the proposed system generates the 3D target facial model by transforming a generic model. Experimental results demonstrate that morphing sequences generated by our systems are of convincing quality.

In path planning, it is often more practical to evaluate the quality of a path not only on the basis of its length but also on the clearance from obstacles. In this paper, we propose computational geometry methods to compute the shortest path with a user-specified minimum clearance between two points in the plane in the presence of disjoint, simple, polygonal obstacles. The algorithm has time complexity O(nlogn) where n is a multiple of the number of obstacle vertices. By setting the minimum clearance to zero, we demonstrate that our algorithm can provide a high quality approximation of the shortest path between source and destination points.

In this simulation we demonstrate how real-time multi-resolution approaches based on underlying geometries are used in terrain visualization. The animated sequence demonstrates applications based on the modified ROAM scheme and the adaptive loop subdivision methods.