REVIEW ON VARIOUS SEGMENTATION TECHNIQUE

Image segmentation is one of the fundamental steps of digital image processing, It is a pre processing step in many algorithms and practical vision system. In image segmentation, digital image divided into multiple set of pixels. There are many algorithms and methods available for image segmentation but still there needs to develop a unique method for it.In this Paper we reviewed on various segmentation technique such as A)Segmentation Based On Edge Detection .B)Region Based Segmentation Method C)Theory Based Segmentation D)Neural Network Based Segmentation E)Model Based Segmentation F)Random Walk Method G)Watershed Segmentation H)Adaptive Mean Shift Algorithm I)Wavelet Based Adaptive Windowing Method I)Vector Quantization J)Level set Mode K)Multi Objective Image Segmentation L)Morphological Segmentation. And it was observed that the results of mathematical morphology based breast tumor segmentation were better. Also segmentation of breast tumor from mammographic images using mathematical morphology was much faster. Hence it can be concluded that mathematical morphology is a better segmentation algorithm for breast tumor demarcation. Segmented results can be further used for feature extraction and classification of breast tumors and we discussed Association Rule Mining Method for Automatic Detection of breast Cancer and it’s work significantly.