Nigel M. Allinson

Characterising Virtual Eigensignatures for General Purpose Face Recognition

We describe an eigenspace manifold for the representation and recognition of pose-varying faces. The distribution of faces in this manifold allows us to determine theoretical recognition characteristics which are then verified experimentally. Using this manifold a framework is proposed which can be used for both familiar and unfamiliar face recognition. A simple implementation demonstrates the pose dependent nature of the system over the transition from unfamiliar to familiar face recognition. Furthermore we show that multiple test images, whether real or virtual, can be used to augment the recognition process. The results compare favourably with reported human face recognition experiments. Finally, we describe how this framework can be used as a mechanism for characterising faces from video for general purpose recognition.

A comprehensive review of current local features for computer vision

Local features are widely utilized in a large number of applications, e.g., object categorization, image retrieval, robust matching, and robot localization. In this review, we focus on detectors and local descriptors. Both earlier corner detectors, e.g., Harris corner detector, and later region detectors, e.g., Harris affine region detector, are described in brief. Most kinds of descriptors are described and summarized in a comprehensive way. Five types of descriptors are included, which are filter-based descriptors, distribution-based descriptors, textons, derivative-based descriptors and others. Finally, the matching methods and different applications with respect to the local features are also mentioned. The objective of this review is to provide a brief introduction for new researchers to the local feature research field, so that they can follow an appropriate methodology according to their specific requirements.

Self-organizing mixture networks for probability density estimation

A self-organizing mixture network (SOMN) is derived for learning arbitrary density functions. The network minimizes the Kullback-Leibler information metric by means of stochastic approximation methods. The density functions are modeled as mixtures of parametric distributions. A mixture needs not to be homogenous, i.e., it can have different density profiles. The first layer of the network is similar to Kohonens self-organizing map (SOM), but with the parameters of the component densities as the learning weights. The winning mechanism is based on maximum posterior probability, and updating of the weights is limited to a small neighborhood around the winner. The second layer accumulates the responses of these local nodes, weighted by the learned mixing parameters. The network possesses a simple structure and computational form, yet yields fast and robust convergence. The network has a generalization ability due to the relative entropy criterion used. Applications to density profile estimation and pattern classification are presented. The SOMN can also provide an insight to the role of neighborhood function used in the SOM.

Multitraining Support Vector Machine for Image Retrieval

Relevance feedback (RF) schemes based on support vector machines (SVMs) have been widely used in content-based image retrieval (CBIR). However, the performance of SVM-based RF approaches is often poor when the number of labeled feedback samples is small. This is mainly due to 1) the SVM classifier being unstable for small-size training sets because its optimal hyper plane is too sensitive to the training examples; and 2) the kernel method being ineffective because the feature dimension is much greater than the size of the training samples. In this paper, we develop a new machine learning technique, multitraining SVM (MTSVM), which combines the merits of the cotraining technique and a random sampling method in the feature space. Based on the proposed MTSVM algorithm, the above two problems can be mitigated. Experiments are carried out on a large image set of some 20 000 images, and the preliminary results demonstrate that the developed method consistently improves the performance over conventional SVM-based RFs in terms of precision and standard deviation, which are used to evaluate the effectiveness and robustness of a RF algorithm, respectively

The yield enhancement of field-programmable gate arrays

The fine granularity and reconfigurable nature of field-programmable gate arrays (FPGAs) suggest that defect-tolerant methods can be readily applied to these devices in order to increase their maximum economic sizes, through increased yield. This paper identifies the inability to contain faults within single cells and the need for fast reconfiguration as the key obstacles to obtaining a significant increase in yield. Monte Carlo defect modeling of the photolithographic layers of VLSI FPGAs is used as a foundation for the yield modeling of various defect-tolerant architectures. Results suggest that a medium-grain architecture is the best solution, offering a substantial increase in size without significant side effects. This architecture is shown to produce greater gate densities than the alternative approach of realizing ultralarge scale FPGAs-multichip modules. >

Image denoising using self-organizing map-based nonlinear independent component analysis

This paper proposes the use of self-organizing maps (SOMs) to the blind source separation (BSS) problem for nonlinearly mixed signals corrupted with multiplicative noise. After an overview of some signal denoising approaches, we introduce the generic independent component analysis (ICA) framework, followed by a survey of existing neural solutions on ICA and nonlinear ICA (NLICA). We then detail a BSS method based on SOMs and intended for image denoising applications. Considering that the pixel intensities of raw images represent a useful signal corrupted with noise, we show that an NLICA-based approach can provide a satisfactory solution to the nonlinear BSS (NLBSS) problem. Furthermore, a comparison between the standard SOM and a modified version, more suitable for dealing with multiplicative noise, is made. Separation results obtained from test and real images demonstrate the feasibility of our approach.

Face recognition from unfamiliar views: subspace methods and pose dependency

A framework for recognising human faces from unfamiliar views is described and a simple implementation of this framework evaluated. The interaction between training view and testing view is shown to compare with observations in human face recognition experiments. The ability of the system to learn from several training views, as available in video footage, is shown to improve the overall performance of the system as is the use of multiple testing images.

On the distribution and convergence of feature space in self-organizing maps

In this paper an analysis of the statistical and the convergence properties of Kohonens self-organizing map of any dimension is presented. Every feature in the map is considered as a sum of a number of random variables. We extend the Central Limit Theorem to a particular case, which is then applied to prove that the feature space during learning tends to multiple gaussian distributed stochastic processes, which will eventually converge in the mean-square sense to the probabilistic centers of input subsets to form a quantization mapping with a minimum mean squared distortion either globally or locally. The diminishing effect, as training progresses, of the initial states on the value of the feature map is also shown.

Proton radiography and tomography with application to proton therapy

Proton radiography and tomography have long promised benefit for proton therapy. Their first suggestion was in the early 1960s and the first published proton radiographs and CT images appeared in the late 1960s and 1970s, respectively. More than just providing anatomical images, proton transmission imaging provides the potential for the more accurate estimation of stopping-power ratio inside a patient and hence improved treatment planning and verification. With the recent explosion in growth of clinical proton therapy facilities, the time is perhaps ripe for the imaging modality to come to the fore. Yet many technical challenges remain to be solved before proton CT scanners become commonplace in the clinic. Research and development in this field is currently more active than at any time with several prototype designs emerging. This review introduces the principles of proton radiography and tomography, their historical developments, the raft of modern prototype systems and the primary design issues.

Advances in self-organising maps

Towards an information-theoretic approach to kernel-based topographic map formation.- A statistical tool to assess the reliability of self-organising maps.- A SOM association network.- A supervised self-organising map for structured data.- Human gait analysis using SOM.- Exploring financial crises data with self-organising maps (SOM).- Analysing health inequalities using SOM.- Integrating contextual information into text document clustering with self-organising maps.- Self-organising Internet semantic network.- Recursive learning rules for SOMs.- Induced Vorono kernels for principal manifolds approximation.- Visualisation induced SOM (ViSOM).- An approach to automated interpretation of SOM.- VQ-based clustering algorithm of piecewise-dependant-data.- Adaptive subspace encoders using stochastic vector quantisers.- SHAPESOM.- A new interpolation algorithm employing a self organising map.- SOM-based exploratory analysis of gene expression data.- Exploring power transformer database using self-organising maps (SOM) and minimal spanning tree (MST).- Recent advances with the growing hierarchical self-organising Map.- Self-organising maps of web link information.- A design method of DNA chips using self-organising maps.- Multi-dimensional self-organising maps on massively parallel hardware.- A new method of Hough transform by using SOM with input vector transformation.- Estimating relevant input dimensions for self-organising algorithms.- An overview on unsupervised learning from data mining perspective.- Recursive self-organising maps.- An investigation into catastrophic interference on a SOM network.- A topography-preserving latent variable model with learning metrics.- Vector quantisation with g-observable neighbors.- Dynamic vector quantisation of speech.- Optimisation of electronic parts mounting machines using SOM-TSP method with 5 dimensional data.- Self-organising maps forcondition assessment of paper insulated cables.- An essay in classifying self-organising maps for temporal sequence processing.- Evaluating SOM-based models in text classification tasks for the Greek language.- Nonlinear blind source separation using SOMs and applications to image denoising.- Signal-based feature extraction and SOM based dimension reduction in a vibration monitoring microsystem.