Search

OCR technology of Latin scripts is well advanced in comparison to other scripts. However, the available results from Latin are not always sufficient to directly adopt them for other scripts such as the Ethiopic script. In this paper, we propose a novel approach that uses structural and syntactic techniques for recognition of Ethiopic characters. We reveal that primitive structures and their spatial relationships form a unique set of patterns for each character. The relationships of primitives are represented by a special tree structure, which is also used to generate a pattern. A knowledge base of the alphabet that stores possibly occurring patterns for each character is built. Recognition is then achieved by matching the generated pattern against each pattern in the knowledge base. Structural features are extracted using direction field tensor. Experimental results are reported, and the recognition system is insensitive to variations on font types, sizes and styles. [ABSTRACT FROM AUTHOR]

We have implemented an algorithm for detection and segmentation of protein spots in 2-D gel electrophoresis images using symmetry derivative features computed using low level image processing operations. The implementation was compared with a previously published Watershed segmentation and a commercial software. Our algorithm was found to yield segmentation results that were either better than or comparable to the other solutions while having fewer free parameters and a low computational cost. [ABSTRACT FROM AUTHOR]

When selecting a registration method for fingerprints, the choice is often between a minutiae based or an orientation field based registration method. In selecting a combination of both methods, instead of selecting one of the methods, we obtain a one modality multi-expert registration system. If the combined methods are based on different features in the fingerprint, e.g. the minutiae points respective the orientation field, they are uncorrelated and a higher registration performance can be expected compared to when only one of the methods are used. In this paper two registration methods are discussed that do not use minutiae points, and are therefore candidates to be combined with a minutiae based registration method to build a multi-expert registration system for fingerprints with expected high registration performance. Both methods use complex orientations fields but produce uncorrelated results by construction. One method uses the position and geometric orientation of symmetry points, i.e. the singular points (SPs) in the fingerprint to estimate the translation respectively the rotation parameter in the Euclidean transformation. The second method uses 1D projections of orientation images to find the transformation parameters. Experimental results are reported. [ABSTRACT FROM AUTHOR]

We present a concise tutorial on statistical learning, the theoretical ground on which the learning from examples paradigm is based. We also discuss the problem of face detection as a case study illustrating the solutions proposed in this framework. Finally, we describe some new results we obtained by means of an object detection method based on statistical hypothesis tests which makes use of positive examples only. [ABSTRACT FROM AUTHOR]

We suggest a novel attentional mechanism for detection of smooth convex and concave objects based on direct processing of intensity values. The operator detects the region of the forearm in images, enabling location of the hand. The operator is robust to variation in illumination, scale, pose, and hand orientation. This method uses the geometrical structure of the forearm, which is common to all people; therefore no limitation of the hand pose and no personal adjustments are required. [ABSTRACT FROM AUTHOR]

Humans are excellent experts in person recognition and yet they do not perform excessively well in recognizing others only based on one modality such as single facial image. Experimental evidence of this fact is reported concluding that even human authentication relies on multimodal signal analysis. The elements of automatic multimodal authentication along with system models are then presented. These include the machine experts as well as machine supervisors. In particular, fingerprint and speech based systems will serve as illustration. A signal adaptive supervisor based on the input biometric signal quality is evaluated. Experimental results on data collected from a mobile telephone prototype application are reported demonstrating the benefits of the reported scheme. [ABSTRACT FROM AUTHOR]

In this paper two methods for human face recognition and the influence of location mistakes are shown. First one, Principal Components Analysis (PCA), has been one of the most applied methods to perform face verification in 2D. In our experiments three classifiers have been considered to test influence of location errors in face verification using PCA. An initial set of "correct located faces" has been used for PCA matrix computation and to train all classifiers. An initial test set was built considering a "correct located faces" set (based on different images than training ones) and then a new test set was obtained by applying a small displacement in both axis (20 pixels) to the initial set. Second method is based on geometrical characteristics constructed with facial and cranial points that come from a 3D representation. Data are acquired by a calibrated stereo system. Classifiers considered for both methods are k-nearest neighbours (KNN), artificial neural networks: radial basis function (RBF) and Support Vector Machine (SVM). Given our data set, results show that SVM is capable to classify correctly in the presence of small location errors. RBF has an acceptable correct rate but the number of false positives is always higher than in the SVM case. [ABSTRACT FROM AUTHOR]

A completely automatic face recognition system is presented. The method works on color and gray level images: after having localized the face and the facial features, it determines 16 facial fiducial points, and characterizes them applying a bank of filters which extract the peculiar texture around them (jets). Recognition is realized measuring the similarity between the different jets. The system is inspired by the elastic bunch graph method, but the fiducial point localization does not require any manual setting. [ABSTRACT FROM AUTHOR]

In this paper, a two-level supervised feature selection algorithm for local feature-based face recognition is presented. In the first part, a genetic algorithm is used to determine the useful locations of the face region for recognition. 2D Gabor wavelet-based feature extractors are used for local image descriptors at these locations. In the second part, the most useful frequencies and orientations of Gabor kernels are determined using a floating feature selection algorithm. Our major aim in this study is to examine the relevance of the two common assumptions in the local feature based face recognition literature: first, that the contribution of a specific feature to the recognition performance is independent of others, and secondly, that feature extractors should be placed over the visually salient points. In this paper, we show that one can obtain better recognition accuracy by relaxing these two assumptions. [ABSTRACT FROM AUTHOR]

In this paper a new complete system for 3D face recognition is presented. 3D face recognition presents several advantages against 2D face recognition, as, for example, invariance to illumination conditions. The proposed system makes use of a stereo methodology, that does not require any expensive range sensors. The 3D image of the face is modelled using Multilevel B-Splines coefficients, that are classified using Support Vector Machines. Preliminary experimental evaluation has produced encouraging results, making the proposed system a promising low cost 3D face recognition system. [ABSTRACT FROM AUTHOR]

Within this paper a technique for model-based 3D hand tracking is presented. A hand model is built from a set of truncated quadrics, approximating the anatomy of a real hand with few parameters. Given that the projection of a quadric onto the image plane is a conic, the contours can be generated efficiently. These model contours are used as shape templates to evaluate possible matches in the current frame. The evaluation is done within a hierarchical Bayesian filtering framework, where the posterior distribution is computed efficiently using a tree of templates. We demonstrate the effectiveness of the technique by using it for tracking 3D articulated and non-rigid hand motion from monocular video sequences in front of a cluttered background. [ABSTRACT FROM AUTHOR]

This tutorial introduces fingerprint recognition systems and their main components: sensing, feature extraction and matching. The basic technologies are surveyed and some state-of-the-art algorithms are discussed. Due to the extent of this topic it is not possible to provide here all the details and to cover a number of interesting issues such as classification, indexing and multi-modal systems. Interested readers can find in [21] a complete and comprehensive guide to fingerprint recognition. [ABSTRACT FROM AUTHOR]

Biometric recognition has attracted the attention of scientists, investors, government agencies as well as the media for the great potential in many application domains. It turns out that there are still a number of intrinsic drawbacks in all biometric techniques. In this paper we postulate the need for a proper data representation which may simplify and augment the discrimination among different instances or biometric samples of different subjects. Considering the design of many natural systems it turns out that spiral (circular) topologies are the best suited to economically store and process data. Among the many developed techniques for biometric recognition, face analysis seems to be the most promising and interesting modality. The ability of the human visual system of analyzing unknown faces, is an example of the amount of information which can be extracted from face images. Nonetheless, there are still many open problems which need to be "faced" as well. The choice of optimal resolution of the face within the image, face registration and facial feature extraction are still open issues. This not only requires to devise new algorithms but to determine the real potential and limitations of existing techniques. In this paper two different methods for face matching are presented, based on the same similarity measure but on different image representations. The methods are tested with conventional and also new databases, obtained from real subjects in real working environments. [ABSTRACT FROM AUTHOR]

This article presents a chapter of "Advanced Studies in Biometrics: Summer School on Biometrics," edited by Massimo Tistarelli, Josef Bigun, and Enrico Grosso, entitled "Biometric Gait Recognition." It examines research into both human and machine recognition of people's gait. Factors such as footwear, terrain, fatigue, injury, and passage of time present difficult challenges.