2.2 - Basic Binary Image Processing

Abstract

This chapter presents the fundamental tools for the processing of binary digital images. Binary image processing is of special interest because an image in binary format can be processed using very fast logical (Boolean) operators. A binary image is obtained from a gray-level image by a process of information abstraction. The simplest abstraction is the process of image thresholding, which can be thought of as an extreme form of gray-level quantization. The process of thresholding is a process of simple comparison: each pixel value in a gray-level image is compared to the gray-scale range. Based on this comparison, a binary decision is made that defines the value of the corresponding pixel in an output binary image. Thresholding is most commonly and effectively applied to images that can be characterized as having bimodal histograms. Bimodal histograms are often associated with images that contain objects and background having significantly different average brightness. This may imply bright objects on a dark background or dark objects on a bright background. The goal, in many applications, is to separate the objects from the background and to label them as object or as background. If the image histogram contains well-separated modes associated with an object and with a background, then thresholding can be the means for achieving this separation.