1 - Vision, the Challenge

Abstract

This chapter discusses some of the problems of machine vision. An obvious means of tackling the recognition problem is to normalize the images in some way. Normalizing the position and orientation of any 2-D picture object would help considerably. Methods for achieving this involve centralizing the objects—arranging that their centroids are at the center of the normalized image—and making their major axes vertical or horizontal. This approach can be taken further if patterns are to be nonrandom, isolated noise points may be eliminated. Ultimately, all these methods help by making the test pattern closer to a restricted set of training set patterns. Computer graphics is the generation of images by computer, starting from abstract descriptions of scenes and knowledge of the laws of image formation. Vision is the process of obtaining descriptions of sets of objects, starting from sets of images and knowledge of the laws of image formation. However, this similarity in formulation of the two processes hides some very fundamental points.