The Research Of Segmentation For VCH-F1 Slice Image

Virtual Human Project, which human further recognize themselves by means of computer and launched, is a significant research topic with wide application value in many fields such as medicine, bionics. Chinese Virtual Human Project is at the visible human stage, namely the moment of reconstruction and understanding of body and tissue organ. The medicine image segmentation method is the key which realizes the Virtual Visual Human stage. slices the image to the colored hypothesized person the organization organ's success division is the core link which the hypothesized visible person plans.This paper, using the chest slice data of Virtual Chinese Human-Female NO.1 (VCH-F1), studied the following methods of medical image segmentation: methods based on boundary, methods based on region and method of uniting the specific theory. Aiming at the characteristic of VCH-F1 image, the segmentation method based on the improvement watershed algorithm and Canny operator are used for gaining the liver, the chest cavity and the heart blood region. The work includes: First,in order to overcome the problem that the edge information of different organs is very fuzzy and use fully R, G, B component gradient information of VCH-F1 color image, the greatest gradient value in the identical coordinate position each component is selected as the gradient value of VCH-F1 slice image.Afterward, in order to eliminate the noise of the gradient image, the iterative multi-threshold segmentation method, considering the difference of gradient image, is used to gain the continuous edge information of the segmentation objects. Then, the methods of region growing and mathematics morphology are used after marking manually object region to gain object template.Next, the edge information of original image gained by Canny operator is used optimizing initial segmentation template for achieving semantic segmentation result. Last, using the relevancy of space position, the shape and the gradation between the goals of neighboring frame, the segmentation result of the preceding image is taken as the reference template of the neighboring frame. The continual automatic segmentation of slice images is realized. Thus the computation is reduced. The segmentation efficiency is enhanced.The experimental results show that this improved method reduces the manual participation, enhances automatic of algorithm, and realizes the VCH-F1 image continuously segmentation accurately.