Wadee Alhalabi

Discriminant Correlation Analysis: Real-Time Feature Level Fusion for Multimodal Biometric Recognition

Information fusion is a key step in multimodal biometric systems. The fusion of information can occur at different levels of a recognition system, i.e., at the feature level, matching-score level, or decision level. However, feature level fusion is believed to be more effective owing to the fact that a feature set contains richer information about the input biometric data than the matching score or the output decision of a classifier. The goal of feature fusion for recognition is to combine relevant information from two or more feature vectors into a single one with more discriminative power than any of the input feature vectors. In pattern recognition problems, we are also interested in separating the classes. In this paper, we present discriminant correlation analysis (DCA), a feature level fusion technique that incorporates the class associations into the correlation analysis of the feature sets. DCA performs an effective feature fusion by maximizing the pairwise correlations across the two feature sets and, at the same time, eliminating the between-class correlations and restricting the correlations to be within the classes. Our proposed method can be used in pattern recognition applications for fusing the features extracted from multiple modalities or combining different feature vectors extracted from a single modality. It is noteworthy that DCA is the first technique that considers class structure in feature fusion. Moreover, it has a very low computational complexity and it can be employed in real-time applications. Multiple sets of experiments performed on various biometric databases and using different feature extraction techniques, show the effectiveness of our proposed method, which outperforms other state-of-the-art approaches.

ResearchGate: An effective altmetric indicator for active researchers?

Abstract As research performance becomes increasingly important for academic institutions in competition for rankings, student recruitment, and funding, many performance indicators have been developed to measure various aspects of research performance. ResearchGate combines bibliometrics and altmetrics to create a more comprehensive performance measure for researchers and institutions. The ResearchGate score, the flagship indicator calculated by an undisclosed algorithm, is a metric that measure scientific reputation. In this research, ResearchGate metrics are firstly compared with those that Research Excellence Framework (REF) and Quacquarelli Symonds (QS) World University Rankings to assess the quality of UK universities and global universities respectively. This study then utilizes correlation analysis to examine whether ResearchGate metrics demonstrate effectiveness on the researcher level in comparison with SciVal metrics. For this research, 300 ResearchGate members from the supply chain management field were selected. The results provide empirical evidence that demonstrate that the ResearchGate score can be an effective indicator for measuring individual researcher performance.

Fully automatic face normalization and single sample face recognition in unconstrained environments

We present a fully automatic face normalization and recognition system.It normalizes the face images for both in-plane and out-of-plane pose variations.The performance of AAM fitting is improved using a novel initialization technique.HOG and Gabor features are fused using CCA to have more discriminative features.The proposed system recognizes non-frontal faces using only a single gallery sample. Single sample face recognition have become an important problem because of the limitations on the availability of gallery images. In many real-world applications such as passport or driver license identification, there is only a single facial image per subject available. The variations between the single gallery face image and the probe face images, captured in unconstrained environments, make the single sample face recognition even more difficult. In this paper, we present a fully automatic face recognition system robust to most common face variations in unconstrained environments. Our proposed system is capable of recognizing faces from non-frontal views and under different illumination conditions using only a single gallery sample for each subject. It normalizes the face images for both in-plane and out-of-plane pose variations using an enhanced technique based on active appearance models (AAMs). We improve the performance of AAM fitting, not only by training it with in-the-wild images and using a powerful optimization technique, but also by initializing the AAM with estimates of the locations of the facial landmarks obtained by a method based on flexible mixture of parts. The proposed initialization technique results in significant improvement of AAM fitting to non-frontal poses and makes the normalization process robust, fast and reliable. Owing to the proper alignment of the face images, made possible by this approach, we can use local feature descriptors, such as Histograms of Oriented Gradients (HOG), for matching. The use of HOG features makes the system robust against illumination variations. In order to improve the discriminating information content of the feature vectors, we also extract Gabor features from the normalized face images and fuse them with HOG features using Canonical Correlation Analysis (CCA). Experimental results performed on various databases outperform the state-of-the-art methods and show the effectiveness of our proposed method in normalization and recognition of face images obtained in unconstrained environments.

The Use of Virtual Reality to Facilitate Mindfulness Skills Training in Dialectical Behavioral Therapy for Borderline Personality Disorder: A Case Study

Borderline personality disorder (BPD) is a severe mental disorder characterized by a dysfunctional pattern of affective instability, impulsivity, and disturbed interpersonal relationships. Dialectical Behavior Therapy (DBT®) is the most effective treatment for Borderline Personality Disorder, but demand for DBT® far exceeds existing clinical resources. Most patients with BPD never receive DBT®. Incorporating computer technology into the DBT® could help increase dissemination. Immersive Virtual Reality technology (VR) is becoming widely available to mainstream consumers. This case study explored the feasibility/clinical potential of using immersive virtual reality technology to enhance DBT® mindfulness skills training of a 32 year old female diagnosed with BPD. Prior to using VR, the patient experienced difficulty practicing DBT® mindfulness due to her emotional reactivity, and difficulty concentrating. To help the patient focus her attention, and to facilitate DBT® mindfulness skills learning, the patient looked into virtual reality goggles, and had the illusion of slowly “floating down” a 3D computer-generated river while listening to DBT® mindfulness training audios. Urges to commit suicide, urges to self harm, urges to quit therapy, urges to use substances, and negative emotions were all reduced after each VR mindfulness session and VR mindfulness was well accepted/liked by the patient. Although case studies are scientifically inconclusive by nature, results from this feasibility study were encouraging. Future controlled studies are needed to quantify whether VR-enhanced mindfulness training has long term benefits e.g., increasing patient acceptance and/or improving therapeutic outcome. Computerizing some of the DBT® skills treatment modules would reduce cost and increase dissemination.

Instance-based ontology matching for e-learning material using an associative pattern classifier

The present work describes a new model of pattern classification and its application to align instances from different ontologies, which are in turn related to e-learning educative content in a Knowledge Society context. In general, ontologies are the fundamental tool inherent to Semantic Web. In particular, the problem of ontology matching is modeled in this paper as a binary pattern classification problem. The original model presented here was validated through experiments, which were done on data taken from the OAEI (Ontology Alignment Evaluation Initiative) 2014 campaign, presented in the OWL (Web Ontology Language) format, as well as on data taken from two international repositories, ADRIADNE and MERLOT, in LOM (Learning Objects Metadata) format. The results obtained show a high precision measurement when compared against some of the best methods present in the state of the art. A new model for ontology matching over educative content repositories is introduced.The model was validated on the OAEI 2014 campaign, ADRIADNE and MERLOT.The homogeneity of resources for e-learning is improved.The results obtained show a high precision measurement.

Discriminant correlation analysis for feature level fusion with application to multimodal biometrics

In this paper, we present Discriminant Correlation Analysis (DCA), a feature level fusion technique that incorporates the class associations in correlation analysis of the feature sets. DCA performs an effective feature fusion by maximizing the pair-wise correlations across the two feature sets, and at the same time, eliminating the between-class correlations and restricting the correlations to be within classes. Our proposed method can be used in pattern recognition applications for fusing features extracted from multiple modalities or combining different feature vectors extracted from a single modality. It is noteworthy that DCA is the first technique that considers class structure in feature fusion. Moreover, it has a very low computational complexity and it can be employed in realtime applications. Multiple sets of experiments performed on various biometric databases show the effectiveness of our proposed method, which outperforms other state-of-the-art approaches.

Virtual reality systems enhance students’ achievements in engineering education

ABSTRACT Virtual reality (VR) is being used for many applications, ranging from medicine to space and from entertainment to training. In this research paper, VR is applied in engineering education, the scope being to compare three major VR systems with the traditional education approach when we do not use any VR system (No-VR). The Corner Cave System (CCS) is compared with the Head Mounted Display (HMD) system. Both of these systems are using a tracking system to reflect the user movements in the virtual environment. The CCS uses only three coordinates: x-, y- and z-axis. The HMD system has six degrees of freedom, the x-, y- and z-axis, as well as the roll, pitch and yaw. Those two systems are also compared with HMD, as a standalone device (HMD-SA) without the tracking system where it has only roll, pitch and yaw. The objective of the study was to evaluate the impact of VR systems on the students’ achievements in engineering colleges. The research examined the effect of the four different methods and compared the scores of the students after each test. The experiments were ran over 48 students. Those systems show incredible results.

From sociability to creditability for academics

Social networking for academic exchanges, such as through sites like Researchgate.net, is gaining popularity among academics. This site offers many metrics (e.g. RG score and RG impact points) which have the potential to become universal research performance metrics. This paper presents an empirical survey of the top 150 researchers grants and their RG scores among 126 colleges of management in Taiwan. Our results show a strong correlation between the research grants and RG scores if the analysis is based on the college as a whole. However, the relationship becomes insignificant for individual researchers. In addition, colleges with the AACSB accreditation outperform AACSB member schools and non-member schools in terms of their research grants and sharing research outputs on ResearchGate. The authors conclude that metrics used on ResearchGate have the potential to become formal research performance evaluation tools. However, this time has not come yet, at least based on the experience of Taiwan. A strong correlation exists between the grants and RG scores at the college level.The relationship becomes insignificant for individual researchers.The AACSB-accredited group outperforms non-AACSB-accredited group.

The use of Virtual Reality facilitates Dialectical Behavior Therapy® “observing sounds and visuals” mindfulness skills training exercises for a Latino patient with severe burns: A case study.

Sustaining a burn injury increases an individuals risk of developing psychological problems such as generalized anxiety, negative emotions, depression, acute stress disorder, or post-traumatic stress disorder. Despite the growing use of Dialectical Behavioral Therapy® (DBT®) by clinical psychologists, to date, there are no published studies using standard DBT® or DBT® skills learning for severe burn patients. The current study explored the feasibility and clinical potential of using Immersive Virtual Reality (VR) enhanced DBT® mindfulness skills training to reduce negative emotions and increase positive emotions of a patient with severe burn injuries. The participant was a hospitalized (in house) 21-year-old Spanish speaking Latino male patient being treated for a large (>35% TBSA) severe flame burn injury. Methods: The patient looked into a pair of Oculus Rift DK2 virtual reality goggles to perceive the computer-generated virtual reality illusion of floating down a river, with rocks, boulders, trees, mountains, and clouds, while listening to DBT® mindfulness training audios during 4 VR sessions over a 1 month period. Study measures were administered before and after each VR session. Results: As predicted, the patient reported increased positive emotions and decreased negative emotions. The patient also accepted the VR mindfulness treatment technique. He reported the sessions helped him become more comfortable with his emotions and he wanted to keep using mindfulness after returning home. Conclusions: Dialectical Behavioral Therapy is an empirically validated treatment approach that has proved effective with non-burn patient populations for treating many of the psychological problems experienced by severe burn patients. The current case study explored for the first time, the use of immersive virtual reality enhanced DBT® mindfulness skills training with a burn patient. The patient reported reductions in negative emotions and increases in positive emotions, after VR DBT® mindfulness skills training. Immersive Virtual Reality is becoming widely available to mainstream consumers, and thus has the potential to make this treatment available to a much wider number of patient populations, including severe burn patients. Additional development, and controlled studies are needed.

Robust multi-scale orientation estimation: Spatial domain Vs Fourier domain

Orientation estimation is considered as an important and vital step towards many pattern recognition and image enhancement tasks. In a noisy environment, the gradient-based estimations provide poor results. A pre-smoothing Gaussian function with an appropriate scale is conventionally used to get better gradients. Later on, fixed-scale approach was extended to include multi-scale gradient estimates. More specifically, multi-scale orientation estimation, based on scale-space axioms, in spatial domain can be formulated. To further boost the performance of multi-scale orientation estimates, a Fourier domain foundation in the form of Directional Filter bank (DFB)is incorporated with multi scale spatial domain approach. This paper presents an approach for estimation of local orientations using multi-scale approach both in spatial and fourier domain. In fourier-domain approach, two linear combinations are deployed, one across the directional image, and the other across scales. This is opposed to only one linear combination across the scales, used in normal spatial domain technique. Simulations are conducted over noisy test images as well as real data. Our objective results indicate that multi-scale fourier domain approach always yields better estimates at variable level of noise as compared to stand alone multi-scale spatial domain. The improvements made by fourier domain estimate can largely be attributed to the use of double linear combination both across directional bands and across scales.