Automatic Detection Method of Leukaemia by Using Segmentation Method

Abstract: In this paper, we are introducing the Disease of Leukaemia or called cancer of blood cell. Number of methods have been introduced for the effectiveness of white blood cell segmentation, multispectral imaging analysis technique like histogram equalization and threshold, k-means clustering, active contours, watershed transform, and many more. These techniques alone not give a significant segmentation rate, that�s why it is better to develop a system which is hybridize with certain segmentation techniques in order to achieve a better segmentation rate in overall procedure. As a new approach we are going to develop a system comprises of techniques like fuzzy c-means clustering, gradient based watershed transform and initial pre-processing done through HSV color transform.

Keywords: [1], HSV color Transform [2], Extract Saturation Component [3], Fuzzy C-means clustering on HSV transformed Image [4], Morphological Operations & Reconstruction Process [5], Gradient Based Watershed Transformation [6], Final Segmented Image (Segmenting leukocytes and blast cells).