Image Thresholding using Generalized Rough Sets - Rough Fuzzy Image Analysis: Foundations and Methodologies - page 57

3.5.1 Qualitative analysis

Segmentation

Let us consider here qualitative assessment of segmentation results in different images in
order to evaluate the performance of various techniques. The techniques considered for
comparison are: (i) the proposed thresholding methodology using the aforesaid grayness
ambiguity measure, (ii) the ambiguity minimization based thresholding method reported in
(Pal et al., 1983) using the aforesaid grayness ambiguity measure (iii) thresholding method
by Otsu (Otsu, 1979), (iv) thresholding method by Kapur et al. (Kapur, Sahoo, and
Wong, 1985), (v) thresholding method by Kittler et al. (Kittler and Illingworth, 1986), (vi)
Thresholding method by Tsai (Tsai, 1985), and (vii) thresholding method by Pal et al. (Pal
et al., 1983). The methods considered will henceforth be referred using their corresponding
numbers.

(a) The Image (b) Graylevel Histogram (c) Segmentation by (i)

(d) Segmentation by (ii) (e) Segmentation by (iii) (f) Segmentation by (iv)

(g) Segmentation by (v) (h) Segmentation by (vi) (i) Segmentation by (vii)

FIGURE 3.8: Performance of the various thresholding algorithms applied to find the core and
extent of the galaxy in an image