Armando Jorge Monteiro Neves Padilha

Linear algebra algorithms in a heterogeneous cluster of personal computers

Cluster computing is presently a major research area, mostly for high performance computing. The work presented refers to the application of cluster computing in a small scale where a virtual machine is composed of a small number of off-the-self-personal computers connected by a low cost network. A methodology to determine the optimal number of processors to be used in a computation is presented as well as the speedup results obtained for the matrix-matrix multiplication and for the symmetric QR algorithm for eigenvector computation which are significant building blocks for applications in the target image processing and analysis domain. The load balancing strategy is also addressed.

Simulation of data distribution strategies for LU factorization on heterogeneous machines

This paper analyzes two static data distributions for inherently sequential algorithms such as the LU factorization, namely the group block distribution and the dynamic programming distribution. The aim is to extend previous comparisons to higher problem and machine dimensions and to obtain guidelines for the right choice for a given problem, that is prior to execution the system should be able to choose the distribution that, for the particular problem at hand, achieves the best performance. Note that, contrary to other algorithms, inherently sequential algorithms present additional difficulties with respect to other classes of algorithms, when the goal is to optimize the processing time, due to the fact that the computational load for data matrix columns increases with their index, requiring a fine tuned load assignment and distribution.

Parallel Image Processing System on a Cluster of Personal Computers

The most demanding image processing applications require real time processing, often using special purpose hardware. The work herein presentedrefers to the application of cluster computing for off line image processing, where the end user benefits from the operation of otherwise idle processors in the local LAN. The virtual parallel computer is composed by off-the-shelf personal computers connected by a low cost network, such as a 10 Mbits/s Ethernet. The aim is to minimise the processing time of a high level image processing package. The system developed to manage the parallel execution is describedand some results obtained for the parallelisation of high level image processing algorithms are discussed, namely for active contour and modal analysis methods which require the computation of the eigenvectors of a symmetric matrix.

Improving Face Recognition by Video Spatial Morphing

The focus of this chapter is in the problem of using technology to grant access to restricted areas by authorised persons, hereafter called ‘clients’, and to deny access to unauthorised or unidentified persons, the so called ‘impostors’. Conventional methods, such as magnetic or smart cards, user/password login and others, are being progressively recognised as insecure due to their many shortcomings, like the possibility of being lost, damaged or forged. Other methods, particularly those based on biometrics, are being increasingly used as they allow the verification of an individual’s identity on the basis of precise and careful measures of biological and physiological characteristics, such as fingerprints, hand and palm print geometry, iris and retina patterns, voice and face recognition. Automatic face recognition has very much progressed in the last few years, making its use practical in experimental or commercial systems. However, further research is still needed to make these systems more robust, reliable and less dependant on special constraints, particularly those imposed on the data acquisition process. In order to be as flexible as possible, current face recognition systems must use a large database of facial views for each client, so that distinct poses and emotional states can be accommodated, as well as other short-term variations in appearance caused by cosmetics or beard size, and by the use of various accessories such as spectacles or earrings. These multiple views are intended to increase the individual’s recognition rate for the capture of a single facial test image. The large dimension of the faces database induces a number of problems, namely the requirement for more storage, the increased computing time for recognition and decision, and the need for more complex classifiers. In an attempt to cope with the above problems we have devised an alternative approach, essentially consisting in keeping a much smaller facial image database, and in testing for valid matches a number of images extracted from a video fragment acquired during the person’s path in direction to the protected entrance. The main advantages to be expected from this approach can be summarised as: (a) the size of the reference face database is substantially reduced, as a single image or a small number of images for each individual are kept, (b) the clients are not subject to much discomfort when building the database, as a single neutral view (for each relevant appearance) is

Algorithm-Dependant Method to Determine the Optimal Number of Computers in Parallel Virtual Machines

Presently computer networks are becoming common in every place, connecting from only a few to hundreds of personal computers or workstations. The idea of getting the most out of the computing power installed is not new and several studies showed that for long periods of the day most of the computers are in the idle state. The work herein refers to a study aiming to find, for a given algorithm, the number of processors that should be used in order to get the minimum processing time. To support the parallel execution the WPVM software was used, under a Windows NT network of personal computers.

Face recognition from spatially-morphed video sequences

The aim of the present work is the recognition of human face visual information, in order to automatically control the access to restricted areas, granting access to authorized “clients” and barring the entrance to “impostors”. The vision system assembled performed the image acquisition, processing and recognition by first creating a database with a single view of each “client” and then by using multiple test images of each individual candidate to access. To get the test images, a video sequence was captured during the individual’s approach path to the camera. Because subjects presented themselves in a random pose before the camera, the synthesis of frontal views was incorporated, by using a view-morphing method. The modelling and the recognition were handled through the use of ICA methods. The identification of valid “clients” was fully successful. In order to check the rejection of “impostors”, a leave-one-out test was performed which gave promising results.While it is generally possible to do recognition from video sequences, the training process is usually done over static images. This is due to the fact that, in many applications (e.g., homeland security), one does not have large video sequences which can be used for training. For example, law enforcement agencies generally have a frontal and a profile view of wanted individuals, but do not usually keep video sequences in file. Nonetheless, in these applications, it is still possible to analyze the information of video sequences for subsequent recognition tasks. This paper presents a probabilistic algorithm that learns from small sets of static images and then recognizes faces from video sequences. The proposed algorithm is robust to partial occlusions, different orientations and expression changes and does not require of precise face localizations. Our preliminary results with a small database show that the proposed method is more robust to such changes than static-to-static recognition of faces.

Experiments on Using WPVM for Industrial Visual Inspection Problems

This paper reports the main results obtained in some experiments conducted in the context of an European project to evaluate the performance and feasibility of using WPVM as a tool to support the parallelisation of application software for industrial visual inspection problems.