ABSTRACT
This research explores a novel approach to multi-modal image fusion, presenting a simultaneous denoising and fusion methodology designed to address the challenges posed by noise in diverse sensor-captured images. The proposed strategy involves the decomposition of noisy source images into cartoon and texture components, where the cartoon component undergoes spatial domain fusion, and the texture component undergoes denoising and fusion using the Block Matching and 3D Filtering (BM3D) model. Experimental results, including SNR, PSNR, and SSI metrics, showcase the consistent superiority of the proposed method over state-of-the-art techniques. The simultaneous denoising and fusion approach not only enhances noise reduction but also preserves image details, leading to enhanced image quality. This innovative methodology holds significant promise for advancing multi-modal image processing applications.
