Research on blood cell image segmentation technology based on neural network method

Abstract

Computer image processing can be used for the processing of microscopic and biomedical images. For example, measuring and diagnosing microscopic images of blood cell smear by computer can improve work efficiency and reduce human error. In order to reduce artificial subjective interference, improve work efficiency and reduce workload, automatic analysis of cell images has become a hot spot in scientific research and clinical application. Among them, automatic segmentation of blood cell images has become one of the most interesting issues. A nonlinear system model based on improved BPNN(BP neural network) is proposed. The experimental results show that it can effectively overcome the unavoidable edge detection problem of existing methods, and make the segmented image closer to the real image.