Urine Sediment Image Segmentation based on Level Set and Mumford-Shah Model

摘要

Because the segmentation of the target with the Level Set is based on the gradient difference of the image while the gradient difference of the blood cells in the image of the urine sediment is not obvious compared with its background. Therefore segmenting it with the Level Set is not satisfactory. The Level Set based on Mumford-Shah model in this paper depends on the comprehensive information of the image, not on the gradient of image. The model is helpful in processing the complicated and the topological structure changing object. This is very important for the segmentation of the complicated topological medical images with less obvious gradient difference. This paper intends to apply the model in the segmentation of the blood cells in the image of the urine sediment. The algorithm of this paper can eliminate the over-segment produced by the Level Set, and obtain an accurate result with fast speed. The experiment shows that this is very effective.