Precise localization and classification method based on iris intestinal ring
Bo Zhang, Chao Zhou · 2023
In order to assess the human intestinal health, precise localization and classification of intestinal rings is required. In this paper, a method based on the user-defined region factor and V-shaped operator is proposed to realize the precise positioning of the inner and outer boundaries of the intestinal ring. Firstly, according to the gray edge information of the inner boundary of the iris intestinal ring, an iterative update positioning method based on four regional factors is proposed to realize the realization of the intestinal ring. The inner boundary is located, and then the outer boundary of the intestinal ring is located according to the improved Gaussian gradient operator. In the process of classification, the texture information of the iris intestinal ring area is the basis for classification, and the healthy and unhealthy iris images are classified by the texture information of the intestinal ring area through the BP network. It is proved by experiments that the accuracy of iris intestinal ring localization in the gallery is 92.67%, and the classification accuracy of healthy and unhealthy iris images is 94.78%