Page 48 -

22 Chapter 2 ■ Edge-Detection Techniques

region represents. While in this example edge detection is merely a step in the
segmentation process, it is sometimes all that is needed, especially when the
objectsinanimage arelines.

1 1

3 3
2 2

(a) (b)
Figure 2.1: Example of edge detection. (a) Synthetic image with blobs on a grey
background. (b) Edge-enhanced image showing only the outlines of the objects.

Consider the image in Figure 2.2, which is a photograph of a cross-section
of a tree. The growth rings are the objects of interest in this image. Each ring
represents a year of the tree’s life, and the number of rings is therefore the same
as the age of the tree. Enhancing the rings using an edge detector, as shown
in Figure 2.2b, is all that is needed to segment the image into foreground
(objects = rings) and background (everything else).

(a) (b) (c)
Figure 2.2: The A cross-section of a tree. (a) Original grey-level image. (b) Ideal edge
enhanced image, showing the growth rings. (c) The edge enhancement that one might
expect using a real algorithm.

Technically, edge detection is the process of locating the edge pixels, and edge
enhancement is the process of increasing the contrast between the edges and
the background so that the edges become more visible. In practice, however, the
terms are used interchangeably, since most edge-detection programs also set