ABSTRACT
Feature extraction is a process that begins with initial feature selection. The selected features will be the major factor that determines the complexity and success of the analysis and pattern classification process. Initially, the features are selected based on the application requirements and the developer’s experience. Feature analysis involves examining the features extracted from the images and determining if and how they can be used to solve the imaging problem under consideration. In some cases, the extracted features may not solve the problem and the information gained by analyzing the features can be used to determine additional features that may be needed. The process is performed iteratively until success is achieved. Here, we discuss shape, histogram, color, Fourier, texture and region-based features. The mathematical concepts of feature vectors and feature hyperspaces are introduced. Feature analysis tools including data normalization, distance and similarity metrics are discussed. Along with the text are 23 illustrative figures and 40 associated monochrome and color images. The end of chapter exercises includes problems and programming exercises.
