Self-learning based image decomposition for blind periodic noise estimation: a dual-domain optimization approach

Periodic noise reduction is a fundamental problem in image processing, which severely affects the visual quality and subsequent application of the data. Most of the conventional approaches are only dedicated to either the frequency or spatial domain. In this research, we propose a dual-domain approach by converting the periodic noise reduction task into an image decomposition problem. We introduced a bio-inspired computational model to separate the original image from the noise pattern without having any a priori knowledge about its structure or statistics. From the filtering perspective, the proposed method filters out only a portion of the noisy frequencies. Some considerations have to be taken into account for computational resources (computing time and memory space) which permits reducing computation complexity without sacrificing the quality of the image reconstruction. In addition, the separator size provided in the decomposition algorithm does not depend on the image size. Experiments on both synthetic and non-synthetic noisy images have been carried out to validate the effectiveness and efficiency of the proposed algorithm. The simulation results demonstrate the effectiveness of the proposed method both qualitatively and quantitatively.