Denoising, Edge Correction, and Enhancement of Breast Cancer Ultrasound Images

Abstract

Mortality rate because of breast cancer diminishes to a large extent if the categorization of breast lesions as malignant or benign is done properly. But this process is quite complicated owing to erroneous detection of noisy pixels as false positives. It can be reduced by proper enhancement of cancer indicating features present in breast cancer ultrasound images. Therefore, the technique proposed for denoising, edge correction, and enhancement is pivoted around two drastic issues. The first issue is related to the blurring of important details because of improper noise suppression, whereas the second issue is associated with poor contrast between background tissues and masses in ultrasound images of breast cancer. In this paper, we propose an ensemble hybrid filter that restores the noisy ultrasound breast cancer images, corrects the edges of restored ultrasound images without getting degraded with additional Gaussian noise which degrades the ultrasound images during edge correction and finally enhances its visual quality. It uses Weiner filtering to remove a small amount of noise, fuzzy derivatives, and smoothing for edge preservation and intensification membership function along with contrast limited adaptive histogram equalization for enhancement. Experimental results are obtained to demonstrate the feasibility of the proposed approach. These results are also compared to traditional Weiner filter by numerical measures and visual inspection.