OTSU Gray-level image segmentation using Otsu s method.
Iseg = OTSU(I,n) computes a segmented image (Iseg) containing n classes
by means of Otsu s n-thresholding method (Otsu N, A Threshold Selection
Method from Gray-Level Histograms, IEEE Trans. Syst. Man Cybern.
9:62-66 1979). Thresholds are computed to maximize a separability
criterion of the resultant classes in gray levels.
OTSU(I) is equivalent to OTSU(I,2). By default, n=2 and the
corresponding Iseg is therefore a binary image. The pixel values for
Iseg are [0 1] if n=2, [0 0.5 1] if n=3, [0 0.333 0.666 1] if n=4, ...
[Iseg,sep] = OTSU(I,n) returns the value (sep) of the separability
criterion within the range [0 1]. Zero is obtained only with images
having less than n gray level, whereas one (optimal value) is obtained
only with n-valued images.
k-means 算法的工作過程說明如下：首先從n個數據對象任意選擇 k 個對象作為初始聚類中心；而對于所剩下其它對象，則根據它們與這些聚類中心的相似度（距離），分別將它們分配給與其最相似的（聚類中心所代表的）聚類；然后再計算每個所獲新聚類的聚類中心（該聚類中所有對象的均值）；不斷重復這一過程直到標準測度函數開始收斂為止。
