White Blood Cell Extraction on Fractional Calculus and Gradient Vector Flow Snake for Leukocyte Classification on Support Vector Machines

Leukocyte is one of the important components in blood cells, and the recognition and classification of leukocytes in an image is significant to the medical diagnosis, but it is a difficult task. In this paper, we studied a new method for the blood cell image preprocessing and segmentation, and for leukocyte classification. In this paper, the blood cell image segmentation and leukocyte classification includes three parts: (1) the most of image noise is removed by using fractional calculus, then the image color information and distance transformation combing the Snake of Gradient Vector Flow (GVF Snake) is applied to extract the leukocytes; (2) then we make the trait of the leukocyte nuclei having higher saturation and utilize a number of morphological operators and the GVF Snake method for detecting the nuclei in the image; and (3) finally, according to the features of the cell colors and textures, we takes advantage of the Support Vector Machines (SVM) to classify leukocytes. The results show that the use of fractional calculus, GVF Snake combining color information for image segmentation, and applying the SVM classifier to classify the leukocytes can achieve satisfactory results.