Edge Detection - Medical Image Processing: Advanced Fuzzy Set Theoretic Techniques - page 178

The size of the median filter is the same as the size of the image window.
For each pixel position, I ( m , n ) is computed and a new difference matrix is
formed. The new matrix is an edge-enhanced image.
If the difference image contains unwanted lines, the image is filtered using
a Gaussian filter. After filtering, the image is then edge detected using stan-
dard Canny's edge detector.

Example 8.2

An example in Figures 8.3 and 8.4 illustrates the effectiveness of intu-
itionistic fuzzy edge detection where the image is initially enhanced
to highlight the structures of the images. Figure 8.3 is a live cell
image, and Figure 8.4 is the nuclei image. It may be noted that the
cells or the nuclei are not clearly visible in the original image. On
enhancing the images, the edges are enhanced and then edge detec-
tion is carried out.

(a) (b) (c)

FIGURE 8.3
(a) Nuclei image, (b) edge-enhanced image using the intuitionistic fuzzy method and (c) edge-
detected image using the intuitionistic fuzzy method. (Modified from Chaira, T., Appl. Soft
Comput. , 12(4), 1259, 2012.)

(a) (b) (c)

FIGURE 8.4
(a) Nuclei image, (b) edge-enhanced image using the intuitionistic fuzzy method and (c) edge-
detected image using the intuitionistic fuzzy method. (Modified from Chaira, T., Appl. Soft
Comput. , 12(4), 1259, 2012.)