Recognition of breast tumor based on gray level co-occurrence matrix and BP neural network

Breast tumor is a kind of woman′s disease with high incidence rate, it is also a kind of disease that can be diagnosed early and treated early. Thus the mortality rate could be reduced. The method of combining grey symbiotic matrix with BP neural network is proposed to improve the recognition rate of breast tumors. Firstly, the infrared breast image was pretreated to highlight the texture of lesions and blood vessels, and the texture features of such gray-scale curve images of the breast were extracted by using the gray co-occurrence matrix. Then, the sample data were trained through BP neural network. The trained BP neural network could effectively identify the lesion area. The experimental results show that the method presented in this paper has good recognition effect on the breast tumor lesion area.