Machine Vision, 3rd Edition

CHAPTER 4 Thresholding Techniques

4.1 Introduction

One of the first tasks to be undertaken in vision applications is to segment objects from their background. When objects are large and do not possess very much surface detail, segmentation is often imagined as splitting the image into a number of regions, each having a high level of uniformity in some parameter such as brightness, color, texture, or even motion. Hence, it should be straightforward to separate objects from one another and from their background, as well as to discern the different facets of solid objects such as cubes.

Unfortunately, this concept of segmentation is an idealization that is sometimes reasonably accurate, but more often in the real world it is an invention of the ...