A number of researchers have investigated the application of neural networks to visual recognition, with much of the emphasis placed on exploiting the network's ability to generalise. However, despite the benefits of such an approach it is not at all obvious how networks can be developed which are capable of recognising objects subject to changes in rotation, translation and viewpoint. In this study, we suggest that a possible solution to this problem can be found by studying aspects of visual psychology and in particular, perceptual organisation. For example, it appears that grouping together lines based upon perceptually significant features can facilitate viewpoint independent recognition. The work presented here identifies simple grouping measures based on parallelism and connectivity and shows how it is possible to train multi-layer perceptrons (MLPs) to detect and determine the perceptual significance of any group presented. In this way, it is shown how MLPs which are trained via backpropagation to perform individual grouping tasks, can be brought together into a novel, large scale network capable of determining the perceptual significance of the whole input pattern. Finally the applicability of such significance values for recognition is investigated and results indicate that both the NILP and the Kohonen Feature Map can be trained to recognise simple shapes described in terms of perceptual significances. This study has also provided an opportunity to investigate aspects of the backpropagation algorithm, particularly the ability to generalise. In this study we report the results of various generalisation tests. In applying the backpropagation algorithm to certain problems, we found that there was a deficiency in performance with the standard learning algorithm. An improvement in performance could however, be obtained when suitable modifications were made to the algorithm. The modifications and consequent results are reported here.

University of Macau / Faculty of Science and Technology. / Department of Computer and Information Science

by Leung Kin Lap. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1996. / Includes bibliographical references (leaves 70-72). / ACKNOWLEDGEMENTS --- p.I / ABSTRACT --- p.II / TABLE OF CONTENTS --- p.III / TABLE OF FIGURES --- p.IV / Chapter Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Contributions of the thesis --- p.2 / Chapter 1.2 --- Organization of the thesis --- p.4 / Chapter Chapter 2 --- Previous Work --- p.5 / Chapter 2.1 --- An overview of 3-D interpretation --- p.5 / Chapter 2.1.1 --- Multiple-View Clues --- p.5 / Chapter 2.1.2 --- Single-View Clues --- p.6 / Chapter 2.2 --- Line Drawing Interpretation --- p.7 / Chapter 2.2.1 --- Qualitative Interpretation --- p.7 / Chapter 2.2.2 --- Quantitative Interpretation --- p.10 / Chapter 2.3 --- Previous Methods of Quantitative Interpretation by Optimization --- p.12 / Chapter 2.3.1 --- Extremum Principle for Shape from Contour --- p.12 / Chapter 2.3.2 --- MSDA Algorithm --- p.14 / Chapter 2.4 --- Comments on Previous Work on Line Drawing Interpretation --- p.17 / Chapter Chapter 3 --- An Iterative Clustering Procedure for Imperfect Line Drawings --- p.18 / Chapter 3.1 --- Shape Constraints --- p.19 / Chapter 3.2 --- Problem Formulation --- p.20 / Chapter 3.3 --- Solution Steps --- p.25 / Chapter 3.4 --- Nearest-Neighbor Clustering Algorithm --- p.37 / Chapter 3.5 --- Discussion --- p.38 / Chapter Chapter 4 --- Experimental Results --- p.40 / Chapter 4.1 --- Synthetic Line Drawings --- p.40 / Chapter 4.2 --- Real Line Drawing --- p.42 / Chapter 4.2.1 --- Recovery of real images --- p.42 / Chapter Chapter 5 --- Conclusion and Future Work --- p.65 / Appendix A --- p.67 / Chapter A. 1 --- Gradient Space Concept --- p.67 / Chapter A. 2 --- Shading of images --- p.69 / Appendix B --- p.70

Abstract With the rapid progress of the MEMs process, the cost and the size of accelerometers are reducing rapidly. As a result, accelerometers have found many new applications in industrial, entertainment and medical domains. One of such an applications is to acquire information about human body movement. The objective of this work is to use knee acceleration signal for indentity verification. Comparing with traditional biometric methods, this approach has several distinct features. First, it can aquire a large amount of data efficiently and conventiently. Second, it is relatively difficult to duplicate. In designing the verification algorithm, this study has developed a neural network method a hyperspherical classifier method. The experimental results demonstrated that hyperspherical classifier provide better performances in this application. By setting the sensitively to 85%, the specificity achieved by the hyperspherical classifier is at least 95%.

Thesis (Ph. D.)--State University of New York at Buffalo, 1998. / "June 1998." Includes bibliographical references (leaves 158-166). Also available in print.

Thesis (Ph.D.)--City University of Hong Kong, 2009. / "Submitted to Department of Computer Science in partial fulfillment of the requirements for the degree of Doctor of Philosophy." Includes bibliographical references (leaves 122-133)

Thesis (Ph. D.) in Physics--University of Maine, 2003. / Includes vita. Includes bibliographical references (leaves 129-143).

Thesis (Ph. D.)--College of Computing, Georgia Institute of Technology, 2004. Directed by Aaron Bobick. / Vita. Includes bibliographical references (leaves 119-123).

Thesis (Ph.D.)--City University of Hong Kong, 2009. / "Submitted to Department of Computer Science in partial fulfillment of the requirements for the degree of Doctor of Philosophy." Includes bibliographical references (leaves 145-161)

An important problem in electromagnetics is that of extracting, interpreting, and exploiting scattering mechanisms from the scattered field of a target. Termed “features”, these physics-based descriptions of scattering phenomenology have many and diverse applications such as target identification, classification, validation, and imaging. In this dissertation, the feature extraction, analysis, and exploitation of both synthetic and measured multidimensional radar signatures are investigated. Feature extraction is first performed on simulated data of the highfrequency electromagnetics solver Xpatch. The scattered, far-field of an electrically large target is well-approximated by a discrete set of points known as scattering centers. Xpatch yields three-dimensional (3D) scattering centers of a target one aspect angle at a time by using the shooting and bouncing ray technique and a computer-aided design (CAD) model of the target. The feature extraction technique groups scattering centers across multiple angles that pertain to the same scattering mechanism. Using a nearest neighbor clustering algorithm, this association is carried-out in a multidimensional grid of scattering center angle, bounce, and spatial location, wherein distinct scattering mechanisms are assumed to be non-overlapping. Synthetic monostatic and bistatic feature sets are extracted and analyzed using this algorithm. Additionally, feature sets are exploited to assist humans in electromagnetic CAD model validation. The generation of target CAD models is a challenging, resource-limited, and human-experience-based process. Target features extracted from a CAD model in question are compared individually to measured data from the physical target by projection of their radar signatures. CAD model disagreements such as missing, added, or dimensionally inaccurate components, as well as measurement imperfections are analyzed. Target traceback information of the features identifies flawed areas of the model. The projection value quantifies the degree of disagreement. The feature extraction methodology is next modified for measured radar signatures which lack readily available scattering center and bounce information. First, many ground plane synthetic aperture radar images of overlapping, limited apertures in azimuth are formed from the measurement data. Then, two-dimensional scattering centers of all images are estimated using a modified CLEAN algorithm. Feature extraction is lastly performed as with Xpatch data, though a reduction in grid dimensionality and orthogonality occurs. Finally, measured feature sets are exploited for sparse elevation 3D imaging and improved CAD model validation. The first application estimates the truth 3D scattering center of each feature using linear least squares to then visualize a composite 3D image of the target. The second application projects both synthetic and measured feature radar signatures to mitigate errors from the intersection of features in the combined measurement signature. / text