Comparative Study on Automated Cell Nuclei Segmentation Methods for Cytology Pleural Effusion Images

Figure 7

Visual cell nuclei segmented using twelve segmentation methods: (a) original image, (b) ground truth image, (c) Otsu’s method, (d) Isodata, (e) maximum entropy, (f) cross entropy, (g) minimum error, (h) fuzzy entropy, (i) adaptive thresholding, (j) k-means clustering, (k) fuzzy c-means clustering, (l) mean shift clustering, (m) Chan–Vese method, and (n) graph-based min cut (note that the images were cropped for better visibility here).

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

(j)

(k)

(l)

(m)

(n)