Fuzzy Morphological Processing Algorithm of Blood Image
Abstract
This article, a morphological image processing algorithm is developed for early detection of viral hepatitis from microscopic blood images. It is mentioned that it can be used to accurately determine each boundary detail in the fuzzy morphological processing of the gray image of the blood image. Boundary detection leads to higher recognition results in neural network training. Compared to the work of a number of scholars, our method offers such performance, but it offers improved dynamics and flexibility in the formulation of linguistic boundary criteria, which can be a leading factor in the design of image processing systems with dynamic and flexible rules, such as Type 2 allows fuzzy rules, which offers an interesting alternative to currently common deep learning applications. The validated experimental results show the reliability of the modified method compared to the Chaira method. Mean, entropy and standard deviation were calculated. It turned out to be higher than the values of the Chaira method.