Automatic White Blood Cell Segmentation based on Image Processing

Authors:	Almuhairy Juma, American University of Sharjah, United Arab Emirates Al-Assaf Yousef, American University of Sharjah, United Arab Emirates
Topic:	8.2 Modelling & Control of Biomedical Systems
Session:	Biomedical Engineering / Biomedical Signal Processing II
Keywords:	Image enhancement, image analysis, image segmentation, medical applications and medical systems.

Abstract

This paper presents the initial work in developing a system to segment and classify white blood cell (WBC) of human peripheral blood smear samples. Segmentation of WBC from other blood elements is achieved using various image processing techniques. Applying thresholding and watershed algorithm on the red channel of the acquired images resulted in producing appropriate segmentation of the blood elements within a smear. Comparing the area, color features of the segmented areas against those of typical WBC singled out the segments that contained WBC for further processing. The proposed segmentation method has given promising segmentation accuracy.