Image Segmentation By Self Organizing Map With Mahalanobis Distance

Image segmentation is the classification of data sets into group of similar data points. This article proposed a method to determine the winner unit by self organizing mapping network. The distance between the input vector and the weight vector has been determined by mahalanobis distance and chooses the unit whose weight vector has the smallest mahalanobis distance from the input vector. The results included in this article show the validity of the proposed method. Keywords—Discriminant function, Image Segmentation, Mahalanobis distance, Self Organizing Map, Unsupervised Neural network.