Adaptive Anisotropic Diffusion Filter in Unsharp Masking Scheme for Mammogram Enhancement using PLIP Operations
Syed Rizwana, Lenin Laitonjam, Ranjita Das · Procedia Computer Science · 2025
Anisotropic diffusion filtering is a noise reduction and edge retention image processing technique based on partial differential equations (PDEs). Unsharp masking is a hybrid filter image-enhancing method that improves contrast and reduces noise while smoothing the edges and texture of mammograms. This work uses Parameterized Logarithmic Image Processing (PLIP) techniques to accomplish unsharp masking. It presents an image-dependent threshold parameter and an adjustable diffusion coefficient. The aim of integrating unsharp masking with adaptive anisotropic diffusion filter image-enhancing techniques is to effectively reduce noise in digital mammograms while preserving brightness and contrast. The diffusion coefficient and threshold parameter choice, however, significantly influence how successful this technique is. Compared to the current methods described in the literature, the suggested method exhibits a significantly faster convergence rate. Comprehensive experiments and analyses show that the new approach outperforms both traditional unsharp masking models and anisotropic diffusion. The outcomes utilizing multiple quality metrics show how effective the suggested methodology is at reducing noise and maintaining edge details. This demonstrates how the process may sharpen tasks, remove noise, and improve the image.