Automatic Grayscale Mapping for Mammograms Enhancement
Jingwei Wang, Junfeng Wang · 2024
Mammography is an effective method for the early diagnosis of breast cancer, so improving the quality of mammograms is critical for accurate diagnosis. This can be achieved through grayscale mapping and high-pass filtering. However, conventional grayscale mapping requires engineers to manually adjust the equipment parameters, which, once set, remain unchanged. This process is not only time-consuming but also fails to consistently ensure image quality. In this paper, we propose an automatic grayscale mapping algorithm. Our method first extracts the grayscale demarcation${\bm{x}}1$between the background and the breast, then identifies the grayscale demarcation${\bm{x}}2$between the breast edges and the breast tissue. Subsequently, a non-linear grayscale mapping based on${\bm{x}}1$and${\bm{x}}2$is conducted to enhance the details of the breast tissue while maintaining the integrity of the breast edges and reducing the background grayscale. After these steps, high-frequency details of the image are further enhanced through high-pass filtering. Experimental results demonstrate the effectiveness and superiority of our method.