Feature Extraction and Image Processing for Computer Vision, 4th Edition

5

High-level feature extraction: fixed shape matching

Abstract

High-level feature extraction concerns finding shapes and objects in computer images. In feature extraction, we generally seek invariance properties so that the extraction result does not vary according to chosen (or specified) conditions. This implies finding objects, whatever their position, their orientation or their size. This chapter concerns shapes that are fixed in shape (such as a segment of bone in a medical image); the following chapter concerns shapes, which can deform (like the shape of a walking person). We first consider whether we can detect objects by thresholding. This is only likely to provide a solution when illumination and lighting can be controlled, so ...