Breast cancer has become a significant world health issue and increasing mortality requires early detection. Although a mammogram is an extremely widely used imaging method for early identification, disturbances, antiques and shifts in the density of tissues frequently impair its quality. Critical traits may be obscured by these distortions which could result in inaccurate findings or incorrect negatives. This study examines how to improve mammography images and increase their quality and accuracy in determining the presence of breast cancer by applying fuzzy low-pass filtering or Gaussian low-pass filtering techniques. As low-pass filters restrict high-frequency peaks while permitting low-frequency signals by, filters are useful for reducing noise. Gaussian filters are frequently used in medical images due to their ease of use and efficacy in decreasing Gaussian noise. However, they frequently fog tiny details, such as important tumour extends and restrictions, which are required for precise diagnosis. Fuzzy low-pass filters, which are based on fuzzy logic principles, provide an alternative filtering strategy that preserves the outermost and fine details of essential structures while significantly decreasing noise. The fuzzy low-pass filtering performed better in terms of PSNR and SSIM compared to Gaussian-based low-pass filtering, based on a comparative study of two techniques aimed at improving mammographic images. The fuzzy filter produces higher PSNR values, indicating better image quality preservation, as well as higher SSIM scores, showing greater structural similarity to the original images. The MSE values for fuzzy filtering are substantially lower showing that it retains more original image data despite removing extraneous noise. The findings imply that fuzzy low-pass filtering is superior for enhancing mammograms especially when diagnosing breast cancer because it maintains diagnostic characteristics like tumour borders and tiny microcalcifications enabling more precise image processing as well as more reputable evaluation inferences.