Design for Embedded Image Processing on FPGAs, 2nd Edition

8Histogram Operations

As indicated in Chapter 7, the parameters for many point operations can be derived from the histogram of pixel values of the image. This chapter is divided into two parts: the first considers greyscale histograms and some of their applications, and the second extends this to multidimensional histograms.

8.1 Greyscale Histogram

The histogram of a greyscale image, as shown in Figure 8.1, gives the count of the number of pixels in an image as a function of pixel value:

(8.1)

Closely related is the cumulative histogram, which counts the number of pixels less than or equal to a pixel value:

(8.2)

The cumulative histogram is always monotonically increasing as is shown in Figure 8.1, with the total number of pixels in the image, .

There are two main steps associated with using histograms for image processing. The first step is to build the histogram, and the second is to extract data from the histogram and use it for processing the image. Building the histogram will be described in Section 8.1.1, whereas the applications are discussed in Sections 8.1.2, 8.1.3, 8.1.4, 8.1.5, 8.1.6.

8.1.1 Building the Histogram

To build the histogram, it is necessary to accumulate ...