Modified Brightness Preserving Bi-Histogram Equalization for Robust Mammogram Image Enhancement

Breast cancer is the most commonly diagnosed cancer in women worldwide. In medical areas where resources are limited, training qualified medical experts to accurately diagnose breast cancer remains a challenge, particularly in digital image processing applications. Brightness preserving bi-histogram equalization (BBHE), a broadly used technique for image improvement, but with a limitation that causes image distortions with severe pixel intensity differences. So, we propose a Modified Brightness Preserving Bi-Histogram Equalization (MBP-BHE) technique by utilizing median intensity for better improvement in predicting feature details of the image. The extensive experimental evaluation was performed on mammogram images from MIAS database. Pixel intensity optimization is utilized to widen the dynamic range for mammogram images to improve color and contrast. The proposed technique is compared to state-of-the-art methods such as Histogram Equalization (HE), Recursive Mean Separation Histogram Equalization (RMSHE), Contrast Limited Adaptive Histogram Equalization (CLAHE), BBHE, and Contrast Stretching. The proposed technique achieves improved outcomes in terms of metrics, i.e., MSE, PSNR, SNR, SSIM, and CNR on test images outperforming in both visual and quantitative terms.