A Hybrid Filtering Technique to Remove Hybrid Noises from Digital Images

Suppression of noises is an immensely important step in image pre-processing. The noises are generated because of the distortion in the sensor’s cell, environmental effects, or transmission error, which can induce artifacts, distortions, or blurring in the images. We aim to suppress the hybrid noise from the distorted images as well as preserve the details and features of the image for improving the efficiency and characteristic of the image processing applications. The most chosen filtering techniques are median and switching-based methods to suppress the artifacts from the images. In this paper, a hybrid non-local means with a switching technique has been introduced, which can suppress the hybrid noise, combinations of impulses, and Gaussian noise, present. The results of the proposed hybrid filter prove its superiority over its predecessors. The quantitative results, PSNR, SSIM, and IEF are showing improvements for the proposed hybrid filter with compared to the state-of-the-art methods as 0.49, 8.46, and 2.53%.