Abstract
Breast cancer is the second highest death causing disease in female. From the data it’s clear that one of the eight women is affected by breast cancer. The most effective way to detect the cancer is by using mammography technique and also it detects masses and abnormal conditions. The cluster shaped white spots on mammograms are used the cancer in the early stages. The importance of breast cancer detection to reduce different noises which are commonly found in mammogram images and provides better image quality. Here the multi-scale and the multi-directional analysis would produce optimal approximation and detail coefficients of Non-subsampled shearlet transform (NSST). Hence in this paper, the noisy image is amended into piecewise smooth function in different subbands of frequency. NSST coefficients are amended into information and noise related coefficients are noise is removed by using the adaptive threshold. To evaluate the performance of this proposed algorithm, Mean Square Error (MSE) and Peak Signal-to-Noise Ratio (PSNR) are used. Testing these experimental results show that the proposed algorithm can preserve the edges and textures very well while weakening the noise can obtain better suppressed noise and enhances the objective of evaluations than other noise removal methods.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have