A normal method about separating leukocytes based on color features

This paper presents an image pre-processing method to separate leukocytes using cytological images, which is an efficient way for detection and classification of leukocytes. Traditional segmentation methods of cytological image are based on monochrome images, which often cause the loss of some useful information. We develop a segmentation method using color features. We find that color segmentation based on hue component in hue, saturation and lightness (HSL) color space is more efficient than monochrome. In our procedure, an operator is employed in the RGB color space to enhance the contrast of the image, achieving the separation of leukocytes according to their histogram distribution of hue component in HSL color space. The results of our experiments show that this new colour feature-based segmentation is far more advantageous than the traditional segmentation methods based on monochrome.