https://www.um.edu.mt/library/oar/handle/123456789/26817 2026-04-04T20:48:35Z 2026-04-04T20:48:35Z https://www.um.edu.mt/library/oar/handle/123456789/27358 2018-02-28T02:35:54Z 2006-01-01T00:00:00Z

Title: How effective are radial basis function neural networks for offline handwritten signature verification? Abstract: The objective of this project was to investigate the effectiveness of totally radial basis function neural network (RBFNN) single-layer architecture for offline handwritten signature verification. An RBFNN, initialised by supervised clustering, was adopted for each author’s signature samples. RBFNNs are quite new in this domain, and are well-known for the robustness in eliminating outliers and for the relatively simple computations required to be trained. These were the main motivator factors that challenged the author of this project to investigate the effectiveness of RBFNNs in the field of offline handwritten signature verification. A signature database was collected for the scope of this study as no international public database is available. Professional recommendations by J. Gaffiero who is a Maltese graphologist and personal recommendations by H. Baltzakis helped to acquire a signature database with as much intrapersonal variations as possible. Three groups of signature features namely global, grid and texture features were used to evaluate the system in different scenarios. The grid and texture features were extracted from a superimposed grid of 12×8 segments, where a vector quantisation (VQ) technique was required to cluster the respective column feature vectors. In this case, two VQ approaches were investigated; an adaptively sized codebook VQ and a fixed size codebook VQ of 50 codewords. The entire system was extensively tested with random signature forgeries and the high recognition rates obtained show that the proposed architecture is effective in this field. Surprisingly, the fixed size codebook VQ performed at least twice as good as the adaptively sized codebook VQ. In fact, the best results where obtained when global and grid features where combined producing a feature vector of 592 elements. In this case a Mean Error Rate (MER) of 2.04% with a False Rejection Rate (FRR) of 1.58% and a False Acceptance Rate (FAR) of 2.5% were achieved. The mentioned results were found to rank better than some other published studies.

2006-01-01T00:00:00Z https://www.um.edu.mt/library/oar/handle/123456789/27043 2018-02-21T02:29:02Z 2008-01-01T00:00:00Z

Title: How effective is a rank-based filter with a frequency and orientation selective response? Abstract: The pioneering study about the functions of the primary visual cortex by Hubel and Wiesel, motivated scientists to design machine vision models which tentatively simulate the same functions. In visual perception, there are two kinds of stimuli, usually referred to as first-order stimuli, characterized by a difference in luminance (the object tends to be lighter or darker than the background) and second-order stimuli that are characterized by difference in texture (the object and background share the same luminance but differ in texture). The literature shows that the acceptable models for first- and second-order stimuli are 2-D Gabor linear filters and the linear-nonlinear-linear (LNL) approach respectively. However, this is still not entirely satisfactory because first- and second-order stimuli are processed through different channels, which require an a priori knowledge of the nature of the image; this is not plausible either from the perceptual or from the application point of view. The existing orientation-selective, nonparametric features, called ranklets, which are based on the computation of Wilcoxon rank-sum test statistics has motivated us to investigate the applicability of other rank statistics which would be suitable for both kinds of stimuli. This study proposes an innovative rank-based filter, with orientation- and frequencyselective response. The filter is based on an approximation of a 2-D Gabor filter and on a combination of the Wilcoxon (location) and the Siegel-Tukey (dispersion) nonparametric statistics. The promising results obtained from experiments on perceptual stimuli show that the proposed filter is sensitive to both first- and second-order stimuli.

2008-01-01T00:00:00Z