ABSTRACT
This chapter focuses on the extraction of image features, which represent various geometric characteristics of shapes in images, such as corners, edges, lines, circles, or ellipses. Image features are used to compare content of images or to determine area of interests. We describe the classical Harris-Stephens and Shi-Tomasi corner detectors and their sub-pixel variants, as well as the usual and Hough transforms for the detection of lines, circles, and ellipses. Finally, we are interested in texture descriptors, and use them to search similar images in large databases.
