Mammographic image enhancement using wavelet transform and homomorphic filter

Abstract

Mammography is the most effective method for the early diagnosis of breast cancer diseases. As mammographic images contain low signal to noise ratio and low contrast, it becomes too difficult for radiologists to analyze mammogram. To deal with the above stated problems, it is very important to enhance the mammographic images using image processing methods. This paper introduces a new image enhancement approach for mammographic images which uses the modified mathematical morphology, wavelet transform and homomorphic filter to suppress the noise of images. For performance evaluation of the proposed method, contrast improvement index (CII) and edge preservation index (EPI) are adopted. Experimental results on mammographic images from Pejvak Digital Imaging Center (PDIC) show that the proposed algorithm improves the two indexes, thereby achieving the goal of enhancing mammographic images.

Keywords

• mammographic images
• image processing
• wavelet transform
• homomorphic filter.