Corner Detection and Matching in Digital Images

The detection of corners in images is essential for many tasks in machine vision, including scene analysis, motion and structure from motion analysis, image registration, image matching, object recognition, robot localization and navigation etc. Corners in an image are the points that show a strong two-dimensional intensity change, and are hence well distinguished from neighboring points. A matching procedure should be followed after the extraction of any local features from an object in order to identify and locate the object. The problem is how to automatically estimate image feature correspondences between two or more images, while at the same time not assigning matches incorrectly. In this book, a comprehensive survey of existing corner detection algorithms is provided. Moreover, an elaborated performance comparison of a set of 14 corner detectors is also included. Then, two novel extensions of existing corner detectors are proposed and finally, two fuzzy set theoretic algorithms for corner matching are presented.