Elizabeth Guest

Robust point correspondence applied to two- and three-dimensional image registration

Accurate and robust correspondence calculations are very important in many medical and biological applications. Often, the correspondence calculation forms part of a rigid registration algorithm, but accurate correspondences are especially important for elastic registration algorithms and for quantifying changes over time. In this paper, a new correspondence calculation algorithm, CSM (correspondence by sensitivity to movement), is described. A robust corresponding point is calculated by determining the sensitivity of a correspondence to movement of the point being matched. If the correspondence is reliable, a perturbation in the position of this point should not result in a large movement of the correspondence. A measure of reliability is also calculated. This correspondence calculation method is independent of the registration transformation and has been incorporated into both a 2D elastic registration algorithm for warping serial sections and a 3D rigid registration algorithm for registering pre and postoperative facial range scans. These applications use different methods for calculating the registration transformation and accurate rigid and elastic alignment of images has been achieved with the CSM method. It is expected that this method will be applicable to many different applications and that good results would be achieved if it were to be inserted into other methods for calculating a registration transformation from correspondences.

A multifaceted active swept millimetre-wave approach to the detection of concealed weapons

The effective detection of concealed handguns and knives in open spaces is a major challenge for police and security services round the world. Here an automated technique for the detection of concealed handguns that relies on active swept illumination of the target to induce both scattered fields and aspect independent responses from the concealed object is presented. The broad frequency sweep permits information about the objects size to be deduced from transformations into the time/distance domain. In our experiments we collect multiple sweeps across the frequency range at very high speed, which produces a time evolved response from the target, from both normal and cross polarized detectors. From this we extract characteristic signatures from the responses that allow those from innocent objects (e.g. mobile phones, keys etc) to be distinguished from handguns. Information about the optical depth separation of the scattering corners and the degree and shape of cross polarization allows a neural network to successfully concealed handguns. Finally this system utilizes a range of signal processing techniques ranging from correlation between cross and normally polarized scattering through to a neural network classifier to deduce whether a concealed weapon is present.

Statistical edge detection of concealed weapons using artificial neural networks

A novel edge detector has been developed that utilises statistical masks and neural networks for the optimal detection of edges over a wide range of image types. The failure of many common edge detection techniques has been observed when analysing concealed weapons X-ray images, biomedical images or images with significant levels of noise, clutter or texture. This novel technique is based on a statistical edge detection filter that uses a range of two-sample statistical tests to evaluate any local image texture differences and by applying a pixel region mask (or kernel) to the image analyse the statistical properties of that region. The range and type of tests has been greatly expanded from the previous work of Bowring et al.1 This process is further enhanced by applying combined multiple scale pixel masks and multiple statistical tests, to Artificial Neural Networks (ANN) trained to classify different edge types. Through the use of Artificial Neural Networks (ANN) we can combine the output results of several statistical mask scales into one detector. Furthermore we can allow the combination of several two sample statistical tests of varying properties (for example; mean based, variance based and distribution based). This combination of both scales and tests allows the optimal response from a variety of statistical masks. From this we can produce the optimum edge detection output for a wide variety of images, and the results of this are presented.

A Neural Network for Counter-Terrorism

This article presents findings concerned with the use of neural networks in the identification of deceptive behaviour. A game designed by psychologists and criminologists was used for the generation of data used to test the appropriateness of different AI techniques in the quest for counter-terrorism. A feed forward back propagation network was developed and subsequent neural network experiments showed on average a 60% success rate and at best a 68% success rate for correctly identifying deceptive behaviour. These figures indicate that, as part of an investigator support system, a neural network would be a valuable tool in the identification of terrorists prior to an attack.

DScentTrail: A New Way of Viewing Deception

The DScentTrail System has been created to support and demonstrate research theories in the joint disciplines of computational inference, forensic psychology and expert decision-making in the area of counter-terrorism. DScentTrail is a decision support system, incorporating artificial intelligence, and is intended to be used by investigators. The investigator is presented with a visual representation of a suspect‟s behaviour over time, allowing them to present multiple challenges from which they may prove the suspect guilty outright or receive cognitive or emotional clues of deception. There are links into a neural network, which attempts to identify deceptive behaviour of individuals; the results are fed back into DScentTrail hence giving further enrichment to the information available to the investigator.