Nikola Pavesic

The complete gabor-fisher classifier for robust face recognition

This paper develops a novel face recognition technique called Complete Gabor Fisher Classifier (CGFC). Different from existing techniques that use Gabor filters for deriving the Gabor face representation, the proposed approach does not rely solely on Gabor magnitude information but effectively uses features computed based on Gabor phase information as well. It represents one of the few successful attempts found in the literature of combining Gabor magnitude and phase information for robust face recognition. The novelty of the proposed CGFC technique comes from (1) the introduction of a Gabor phase-based face representation and (2) the combination of the recognition technique using the proposed representation with classical Gabor magnitude-based methods into a unified framework. The proposed face recognition framework is assessed in a series of face verification and identification experiments performed on the XM2VTS, Extended YaleB, FERET, and AR databases. The results of the assessment suggest that the proposed technique clearly outperforms state-of-the-art face recognition techniques from the literature and that its performance is almost unaffected by the presence of partial occlusions of the facial area, changes in facial expression, or severe illumination changes.

A Fast Simplified Fuzzy ARTMAP Network

We present an algorithmic variant of the simplified fuzzy ARTMAP (SFAM) network, whose structure resembles those of feed-forward networks. Its difference with Kasubas model is discussed, and their performances are compared on two benchmarks. We show that our algorithm is much faster than Kasubas algorithm, and by increasing the number of training samples, the difference in speed grows enormously.The performances of the SFAM and the MLP (multilayer perceptron) are compared on three problems: the two benchmarks, and the Farsi optical character recognition (OCR) problem. For training the MLP two different variants of the backpropagation algorithm are used: the BPLRF algorithm (backpropagation with plummeting learning rate factor) for the benchmarks, and the BST algorithm (backpropagation with selective training) for the Farsi OCR problem.The results obtained on all of the three case studies with the MLP and the SFAM, embedded in their customized systems, show that the SFAMs convergence in fast-training mode, is faster than that of MLP, and online operation of the MLP is faster than that of the SFAM. On the benchmark problems the MLP has much better recognition rate than the SFAM. On the Farsi OCR problem, the recognition error of the SFAM is higher than that of the MLP on ill-engineered datasets, but equal on well-engineered ones. The flexible configuration of the SFAM, i.e. its capability to increase the size of the network in order to learn new patterns, as well as its simple parameter adjustment, remain unchallenged by the MLP.

Personal recognition based on an image of the palmar surface of the hand

This paper describes the design and development of a multimodal biometric personal recognition system based on features extracted from a set of 14 geometrical parameters of the hand, the palmprint, four digitprints, and four fingerprints. The features are extracted from a single high-resolution gray-scale image of the palmar surface of the hand using the linear discriminant analysis (LDA) appearance-based feature-extraction approach. The information contained in the extracted features is combined at the matching-score level. The resolutions of the palmprint, digitprint and fingerprint sub-images, the similarity/dissimilarity measures, the matching-score normalization technique, and the fusion rule at the matching-score level, which optimize the system performance, were determined experimentally. The biometric system, when using a system configuration with optimum parameters, showed an average equal error rate (EER) of 0.0005%, which makes it sufficiently accurate for use in high-security biometric systems.

An Evaluation of Video-to-Video Face Verification

Person recognition using facial features, e.g., mug-shot images, has long been used in identity documents. However, due to the widespread use of web-cams and mobile devices embedded with a camera, it is now possible to realize facial video recognition, rather than resorting to just still images. In fact, facial video recognition offers many advantages over still image recognition; these include the potential of boosting the system accuracy and deterring spoof attacks. This paper presents an evaluation of person identity verification using facial video data, organized in conjunction with the International Conference on Biometrics (ICB 2009). It involves 18 systems submitted by seven academic institutes. These systems provide for a diverse set of assumptions, including feature representation and preprocessing variations, allowing us to assess the effect of adverse conditions, usage of quality information, query selection, and template construction for video-to-video face authentication.

Training RBF networks with selective backpropagation

Backpropagation with selective training (BST) is applied on training radial basis function (RBF) networks. It improves the performance of the RBF network substantially, in terms of convergence speed and recognition error. Three drawbacks of the basic backpropagation algorithm, i.e. overtraining, slow convergence at the end of training, and inability to learn the last few percent of patterns are solved. In addition, it has the advantages of shortening training time (up to 3 times) and de-emphasizing overtrained patterns. The simulation results obtained on 16 datasets of the Farsi optical character recognition problem prove the advantages of the BST algorithm. Three activity functions for output cells are examined, and the sigmoid activity function is preferred over others, since it results in less sensitivity to learning parameters, faster convergence and lower recognition error.

Adaptation of SIFT features for robust face recognition

The Scale Invariant Feature Transform (SIFT) is an algorithm used to detect and describe scale-, translation- and rotation-invariant local features in images. The original SIFT algorithm has been successfully applied in general object detection and recognition tasks, panorama stitching and others. One of its more recent uses also includes face recognition, where it was shown to deliver encouraging results. SIFT-based face recognition techniques found in the literature rely heavily on the so-called keypoint detector, which locates interest points in the given image that are ultimately used to compute the SIFT descriptors. While these descriptors are known to be among others (partially) invariant to illumination changes, the keypoint detector is not. Since varying illumination is one of the main issues affecting the performance of face recognition systems, the keypoint detector represents the main source of errors in face recognition systems relying on SIFT features. To overcome the presented shortcoming of SIFT-based methods, we present in this paper a novel face recognition technique that computes the SIFT descriptors at predefined (fixed) locations learned during the training stage. By doing so, it eliminates the need for keypoint detection on the test images and renders our approach more robust to illumination changes than related approaches from the literature. Experiments, performed on the Extended Yale B face database, show that the proposed technique compares favorably with several popular techniques from the literature in terms of performance.

Illumination invariant face recognition by non-local smoothing

Existing face recognition techniques struggle with their performance when identities have to be determined (recognized) based on image data captured under challenging illumination conditions. To overcome the susceptibility of the existing techniques to illumination variations numerous normalization techniques have been proposed in the literature. These normalization techniques, however, still exhibit some shortcomings and, thus, offer room for improvement. In this paper we identify the most important weaknesses of the commonly adopted illumination normalization techniques and presents two novel approaches which make use of the recently proposed non-local means algorithm. We assess the performance of the proposed techniques on the YaleB face database and report preliminary results.

Principal Gabor filters for face recognition

Gabor filters have proven themselves to be a powerful tool for facial feature extraction. An abundance of recognition techniques presented in the literature exploits these filters to achieve robust face recognition. However, while exhibiting desirable properties, such as orientational selectivity or spatial locality, Gabor filters have also some shortcomings which crucially affect the characteristics and size of the Gabor representation of a given face pattern. Amongst these shortcomings the fact that the filters are not orthogonal one to another and are, hence, correlated is probably the most important. This makes the information contained in the Gabor face representation redundant and also affects the size of the representation. To overcome this problem we propose in this paper to employ orthonormal linear combinations of the original Gabor filters rather than the filters themselves for deriving the Gabor face representation. The filters, named principal Gabor filters for the fact that they are computed by means of principal component analysis, are assessed in face recognition experiments performed on the XM2VTS and YaleB databases, where encouraging results are achieved.

Information systems security and human behaviour

Until recently, most of the effort for providing security in information systems has been focused on technology. However, it turned out during the last years that human factors have played a central role. Therefore, to ensure appropriate security in contemporary information systems, it is necessary to address not only technology-related issues, but also human behaviour and organisation-related issues that are usually embodied in security policies. This paper presents a template model, which is intended to support risk management for information systems, and which is concentrated on human factors. The model is based on business dynamics that provide the means for qualitative and quantitative treatment of the above-mentioned issues.

Gray level thresholding using the Havrda and Charvat entropy

Investigating the Kapur et al. (1985) image thresholding method, we found, that taking the sum of the Havrda and Charvat entropies as a criterion for threshold selection instead of the Shannon entropies, can result in a better image segmentation in the sense of greater uniformity of the partitioned segments, as well as greater contrast among segments.