A modified NLM method for noise remove based on sequential images

Abstract

This paper proposes an improved Non-Local Means denoising method based on multiple images. The main idea of this method is to use several sequential images to improve the denoising performance of NLM algorithm. This method not only takes into consideration the self-similarity of images, but also uses the similarity between sequential images. This algorithm has been verified by using synthetic images with different levels of noise and real images. PSNR and SSIM has been introduced to evaluate the quality of images after processing. Experiments show that this algorithm is able to remove the noise, and retain the details of images at the same time.