IMAGE SEGMENTATION USING NEURO-FUZZY CLUSTERING TECHNIQUE

Abstract

In the recent years, colour image segmentation has attracted more and more attention, this comes of the usefulness or even the necessity of the colour for pattern recognition and computer vision. One of the most important operations in Computer Vision is segmentation. The aim of image segmentation is the domain-independent partition of the image into a set of regions which are visually distinct and uniform with respect to some property, such as grey level, texture or colour. The problem of segmentation has been, and still is, an important research field and many segmentation methods have been proposed in the literature. In this work k-means, fuzzy c-means, fuzzy kohenon and adaptive neuro-fuzzy clustering techniques are used for the colour image segmentation. The most common clustering technique is the K-means algorithm. It is known as a powerful method to deal with the large color pixel set to get the optimal clustering. In this scheme, each pixel is mapped to the one point in the feature space according to its feature (typically color). The feature points that having similar feature are grouping into the same cluster. Then, the each feature point has the cluster index and it is inversely mapped to the image space. One of the well-known unsupervised algorithms that have been applied to many applications is fuzzy c-mean (FCM) other examples exist alike, among which we site: PCM (Probabilistic C-Means), K-nearest neighbour, and K-means. Compared with crisp or hard segmentation methods, FCM is able to retain more information from the original image. Fuzzy Kohonen clustering network which integrates the Fuzzy c-Means (FCM) model into the learning rate and updating strategies of the Kohonen network. This yields an optimization problem related to FCM. An auto adaptive neuro-fuzzy segmentation. The system consists of a multilayer perceptron (MLP)-like network that performs image segmentation by adaptive thresholding of the input image using labels automatically pre-selected by a fuzzy clustering technique. The proposed architecture is feed forward, but unlike the conventional MLP the learning is unsupervised. iii The output status of the network is described as a fuzzy set. Fuzzy entropy is used as a measure of the error of the segmentation system .The proposed system is capable to perform automatic multilevel segmentation of images, based solely on information contained by the image itself.