Segmentation of Colour Data Base Image by Implementing K-Means Clustering

Abstract:In image analysis techniques, image segmentation takes a major role for analyzing any type of image. The Kmeans clustering algorithm is one of the widely used algorithm in image segmentation system. This paper proposes the colour data base image segmentation using the L*a*b* colour space and K-means clustering. This work presents a data base image segmentation based on colour features with K-means clustering unsupervised algorithm developed with MATLAB coding. The entire work is divided into two stages. First enhancement of colour separation of data base colour image using de-correlation stretching is carried out and then the six data base image regions are grouped into a set of three clusters using K-means clustering algorithm. By applying the L*a*b* colour space and k-means clustering algorithm in colour data base image we can only point out the major area of any image. By this process we can isolate the infected area in medical data base colour image and cure the disease easily. For better result we can use some optimization technique like Particle swarm Optimization(PSO).