Texture segmentation by frequency-sensitive elliptical competitive learning

In this paper, a new learning algorithm is proposed with the purpose of tex ture segmentation. The algorithm is a competitive clustering scheme with tw o specific features: elliptical clustering is accomplished by incorporating the Mahalanobis distance measure into the learning rules and under-utiliza tion of smaller clusters is avoided by incorporating a frequency-sensitive term. In the paper, an efficient learning rule that incorporates these feat ures is elaborated. In the experimental section, several experiments demons trate the usefulness of the proposed technique for the segmentation of text ured images. On the compositions of textured images, Gabor filters were app lied to generate texture features. The segmentation performance is compared to k-means clustering with and without the use of the Mahalanobis distance and to the ordinary competitive learning scheme. It is demonstrated that t he proposed algorithm outperforms the others. A fuzzy version of the techni que is introduced, and experimentally compared with fuzzy versions of the k -means and competitive clustering algorithms. The same conclusions as for t he hard clustering case hold. (C) 2001 Elsevier Science B.V. All rights res erved.