The segmentation of image is the basic thing for understanding the images whether it is a color image or gray scale image. It is used in the various image processing applications, computer vision, etc. In this thesis work we have used multiple clustering approaches to segment the image in our initial step like Normalized cut, kMeans, and Mean shift. The main aim was to obtain feature extraction, to reduce convergence, to reduce computation time, and to overcome the over segmentation caused by the noise, also incorrect spread of intensity. Hence the optimal solution has been derived through the Modified kMeans through which the feature extraction and the separation of overlapping objects were evaluated by making use of wavelet transform and computation time was reduced by considering approximation band coefficients of DWT contribution in an image through which overall performance was improved. Proposed work has been implemented in MATLAB environment.
Keywords
Image Segmentation, Normalized Cut, Mean shift, kMeans, Modified kMeans.
