BASICS OF THRESHOLD CONVERSION OF BRIGHTNESS IN COLOR IMAGES

Abstract

Threshold transformations occupy a central place in applied problems of image segmentation due to intuitive properties, ease of implementation, and computational speed. In this section, a more formal definition of the threshold transformation is given and on its basis much more general methods of threshold processing are constructed than are described so far. Suppose that shown in fig. 2.3.1. (a) the histogram corresponds to some image containing bright objects against a dark background, so that the brightness of the pixels of the object and background are concentrated near the two prevailing values. The obvious way to distinguish objects from the surrounding background is to select a threshold value , delimiting brightness distribution modes. Then any point wherein called the point of the object, otherwise, the point of the background