Aumomatic classification of cells using morphological shape in peripheral blood images

摘要

A novel techniqe for automatic analysis and classification of celkls in peripheral blood images is presented. The puposes of this research are to analyze and classify morphological shapes of mature red-blood cells and whilte-blood cells in peripheral blood images. We first, identify red-blood cells and white-blood cells in a blood cells in a blood image capatured from CCD camera attached to microscope. Feature extraction is the second step. Finally blood cells are classified using back propagation neural network. Fifteen different classification clusters including normal cells are in red blood cell. However, there are five different normal categories in discrimination of white boood cells. In other words, the system can tell whether a given white cell belongs to a one of vive normal classes or not. A novel segmentation method is presented for extraction of nucleus and cytoplasma which inherently posses valuable cues in white blood cell classification. Initially, seventy-six dimensions of a feature vector that includes UNL Fourier descriptor, shape, and color are considered in red-blood cell classification. While, thirty-eight dimensions of a feature vector are considered in red blood cell classification Based on the proposed method, a prototpe system has implemenmted and evaluated with various classification algorithms such as LVQ-3 (Learning Vecgtor Quantization) and K-NN (K-nearest nighbor). The experiment results show that the proposed method out performs on blood cell classification compared with other alternatives.