Red Blood Cell classification through shape feature extraction and PSO-CSVM classifier design

The automatic procession of erythrocyte image is helpful to clinic blood-related disease treatment in Medical Image Computer Aided Diagnosing MICAD. The original input data we concerned were Red Blood Cell images captured by Scanned Electron Microscope(SEM). After 3D height field recovered from the varied shading, the depth map of each point on the surfaces is applied to calculate Gaussian curvature and mean curvature, which are used to produce surface type label image. Accordingly the surface is segmented into different parts through multi-scale bi-variate polynomials function fitting. The count of different surface types is used to design a classifier for training and classifying the red blood cell by means of support vector machine and particle swarm optimization. The combined classifier shows efficient and easily to be implemented.