An adaptive leukocyte nucleus segmentation using genetic algorithm

Leukocyte segmentation and location detection is the most important preprocessing step for further recognition. In this paper, an adaptive leukocyte segmentation method is proposed. Two kinds of color spaces are considered to enhance the nuclei. With the combined color space, the variety of stain and light condition can be avoided. In order to deal with different sizes of images, an adaptive segmentation method based on genetic algorithm is proposed. The experimental result shows that we can obtain promised segmental results even apply on different color tone and size of smear images.