Artificial Intelligence in Recognition and Classification of Astrophysical and Medical Images

During the past decade digital imaging has significantly progressed in all imaging areas ranging from medicine, pharmacy, chemistry, biology to astrophysics, meteorology and geophysics. The avalanche of digitized images produced a need for special techniques of processing and knowledge extraction from many digital images with minimal or even without human interaction. This has resulted in a new area in the digital processing called pattern recognition that becomes increasingly necessary owing to a growing number of images to be processed. The first applications of pattern recognition techniques were for the analysis of medical X rays and MMR images that enabled the extraction of quantified information in terms of texture, intensity and shape and allowed to significantly improve a di agnosis of human organs. These techniques were significantly developed over the la st few years and combined feature detection and classification by using re gion based and artificial intelligence methods. By using growing databases of medical images processed with pattern recognition and classification t echniques, one can produce fast and consistent diagnosis of diseases based on the accumulated knowledge obtained from many other similar cases from the stored databases. The use of CCD cameras for astroph ysical instruments on the ground and space produce digitized images in va rious fields of astrophysics. In the past decade, many space and ground based instruments provide large numbers of digitized images of the ni ght skies and of the Sun, our closest star.

Autorentext
Valentina Zharkova is Chair of Applied Mathematics at the Department of Mathematics and Computing, University of Bradford, UK. Having obtained her M.Sc. degree from Taras Shevchenko National University of Kiev (NUK), Ukraine, and her Ph.D. degree from the Main Astronomical Observatory of the Ukrainian Academy of Sciences, she worked for NUK for 20 years. In 1992, she moved to the University of Glasgow as a Royal Society Research Fellow and in 2000 to the University of Bradford. She authored over 200 scientific publications including the key one in Nature (issue 6683, 1998) on the discovery of solar quakes, which was covered by media worldwide. Professor Zharkova is a member of the Royal Astronomical Society, the International Astronomical Union, the European Physical Society and of the Editorial Board of the Journal on Knowledge-based Engineering Systems (KES).

Klappentext
This book presents innovative techniques in Recognition and Classification of Astrophysical and Medical Images. The contents include: Introduction to pattern recognition and classification in astrophysical and medical images. Image standardization and enhancement. Region-based methods for pattern recognition in medical and astrophysical images. Advanced information processing using statistical methods. Feature recognition and classification using spectral method The book is intended for astrophysicists, medical researches, engineers, research students and technically aware managers in the Universities, Astrophysical Observatories, Medical Research Centres working on the processing of large archives of astrophysical or medical digital images. This book can be used as a text book for students of Computing, Cybernetics, Applied Mathematics and Astrophysics. While there are plenty of volumes tackling pattern recognition problems in finance, marketing, and the like, I commend the editors and the authors for their efforts to tackle the big questions in life, and their excellent contributions to this book. Professor Kate Smith-MilesHead, School of Engineering and Information Technology, Deakin University, Australia

Inhalt
to Pattern Recognition and Classification in Medical and Astrophysical Images.- Image Standardization and Enhancement.- Intensity and Region-Based Feature Recognition in Solar Images.- Advanced Feature Recognition and Classification Using Artificial Intelligence Paradigms.- Feature Recognition and Classification Using Spectral Methods.