UNSUPERVISED SEGMENTATION OF MICROSCOPIC WHITE BLOOD CELLS IMAGES USING HISTOGRAM EQUALIZATION TECHNIQUE

Abstract

In this paper, a new method is presented to segment the microscopic images of White Blood Cells (WBC) into cell and non-cell regions depending on the image histogram. The proposed method involves of: first, separating the captured image into its three color channels (Red, Green, and Blue) to produce three grayscale files one for each. Second, apply the histogram equalization upon the three files to obtain the white blood cells image to be segmented into white background, black nucleus with color stains inside and surrounded by cytoplasm. Experimental results show that the proposed unsupervised segmentation method have given reliable results comparable with thresholding method.