In automatic signature verification, questioned specimens are usually compared with reference signatures. In writer-dependent schemes, a number of reference signatures are required to build up the individual signer model while a writer-independent system requires a set of reference signatures from several signers to develop the model of the system. This paper addresses the problem of automatic signature verification when no reference signatures are available. The scenario we explore consists of a set of signatures, which could be signed by the same author or by multiple signers. As such, we discuss three methods which estimate automatically the common authorship of a set of off-line signatures. The first method develops a score similarity matrix, worked out with the assistance of duplicated signatures; the second uses a feature-distance matrix for each pair of signatures; and the last method introduces pre-classification based on the complexity of each signature. Publicly available signatures were used in the experiments, which gave encouraging results. As a baseline for the performance obtained by our approaches, we carried out a visual Turing Test where forensic and non-forensic human volunteers, carrying out the same task, performed less well than the automatic schemes.

This paper describes an application of multi-criteria decision making (MCDM) for multi-modal fusion of features in a 3D face recognition system. A decision making process is outlined that is based on the performance of multi-modal features in a face recognition task involving a set of 3D face databases. In particular, the fuzzy interval valued MCDM technique called TOPSIS is applied for ranking and deciding on the best choice of multi-modal features at the decision stage. It provides a formal mechanism of benchmarking their performances against a set of criteria. The technique demonstrates its ability in scaling up the multi-modal features.

In this paper, the author presents a work on i) range data and ii) stereo-vision system based disparity map profiling that are used as signatures for 3D face recognition. The signatures capture the intensity variations along a line at sample points on a face in any particular direction. The directional signatures and some of their combinations are compared to study the variability in recognition performances. Two 3D face image datasets namely, a local student database captured with a stereo vision system and the FRGC v1 range dataset are used for performance evaluation.

This work describes the process of face synthesis by image morphing from less expensive 3D sensors such as KIECT that are prone to sensor noise. Its main aim is to create a useful face database for future face recognition studies.

In this paper, edge detection techniques and their performance are compared when applied in license plate detection using an embedded digital signal processor. License plate detection remains to be the crucial part of a vehicle's license plate recognition process. The edge detection algorithms compared in this work are those reported capable of delivering real-time performance. These are Canny-Deriche-FGL, Haar and Daubechies-4 wavelet transform and the classic Sobel. These particular algorithms are chosen and compared due to their good performance on digital signal processors. The comparison is drawn in terms of speed and detection success of a license plate. The results show Haar wavelet-based edge detector performs better on a DSP with LP detection speed of 7.32 ms and 98.6% success using 45,032 UK images containing license plates at 768×288 resolutions.

The detection of license plate region is the most important part of a vehicle's license plate recognition process followed by plate segmentation and optical character recognition. Edge detection is commonly used in license plate detection as a preprocessing technique. This paper compares the performance of the image enhancement filters when used in edge detection algorithms combined with connected component analysis to extract license plate region. The experimental comparison of Canny, Kirsch, Rothwell, Sobel, Laplace and SUSAN edge detectors on gray scale images shows that Canny yields high plate detection of 98.2% tested on 45,032 UK images containing license plates at 720×288 resolution captured under various illumination conditions. The average processing time of one image is 56.4 ms.

Automatic Number Plate Recognition (ANPR) systems allow users to track, identify and monitor moving vehicles by automatically extracting their number plates. This paper presents an improved method to locate car plates in an ANPR system. The proposed method is based on morphological open and close operations where different Structuring Elements (SE) are used to maximally eliminate non-plate region and enhance plate region. This method has been tested using a database of UK number plates and results achieved have shown significant improvements in terms of the detection rate compare to other existing plate localisation systems.

In this paper, a fuzzy surface classification paradigm, which is an extension to the conventional techniques based on the sign of the mean (H) and Gaussian (K) curvatures, respectively is presented. With the conventional methods, two of the major problems that limit object descriptions are: 1) Their inability to describe surfaces in a natural way, and 2) computation of curvatures being highly sensitive to noise as well as limited by resolution. Problem 1) is addressed by treating the transitional regions between distinct surface types as smoothly varying (fuzzy) surface types. Problem 2) gets partially resolved while fuzzifying the signs of the surface curvatures for surface description. The new segmentation technique is demonstrated in a model-based object recognition system and its performance is compared with a system based on conventional surface classification.