Denoising of natural images through robust wavelet thresholding and genetic programming

Abstract

Digital images play an essential role in analysis tasks that can be applied in various knowledge domains, including medicine, meteorology, geology, and biology. Such images can be degraded by noise during the process of acquisition, transmission, storage, or compression. The use of local filters in image restoration may generate artifacts when these filters are not well adapted to the image content as a result of the heuristic optimization of local filters. Denoising methods based on learning procedure are more capable than parametric filters for addressing the conflicts between noise suppression and artifact reduction. In this study, we present a nonlinear filtering method based on a two-step switching scheme to remove both salt-and-pepper and additive white Gaussian noises. In the switching scheme, two cascaded detectors are used to detect noise, and two corresponding estimators are employed to effectively and efficiently filter the noise in an image. In the process of training, a method according to patch clustering is utilized, and genetic programming (GP) is subsequently applied to determine the optimum filter (wavelet-domain filter) for each individual cluster, while B Asem Khmag in testing part, the optimum filter trained beforehand by GP is recovered and used on the inputted corrupted patch. This adaptive structure is employed to cope with several noise types. Experimental and comparative analysis results show that the denoising performance of the proposed method is superior to that of existing denoising methods as per both quantitative and qualitative assessments.

Natural Digital image Mixed Noise Removal Using Regularization Perona-Malik Model and Pulse Coupled Neural Networks

“…All the experiments conducted in this paper are run on a 2.2 GHz Intel Core i7 processor with 8GB 1600 MHz DDR3 memory using MATLAB (R2018b). As depicted in figure 2, 3, 4 and 5 benchmark digital images of Couple, Cars, Lena and Jar, (a) is the original image, (b) represents the noisy image with 20% impulse noise density, (c) denoised image using NL-mean [11] (d) denoised image using deep learning-based methods such as DnCNN [23] (e) denoised image using weighted nuclear norm minimization (WNNM) [9] and finally denoised image using proposed method is depicted in (f). As can be seen from several benchmark images, different denoising techniques have certain inhibitory influence for impulse in the contaminated images, but each the algorithm of denoising effect is not identical.…”

“…In this method, the mixed and Gaussian noise types are suppressed. Another study that exploited Genetic programing is proposed by [11]. The additive Gaussian noise are located based on the synchronization pulse distribution where the characteristics of the pulse contaminated coupled in clusters to ease remove the noise and filter the corrupted pixels.…”

Genetic Programming to Remove Impulse Noise in Color Images

“…On the other hand, genetic programming has emerged as a highly promising field for the creation of adaptive filters aimed at noise reduction; however, no filters use genetic programming to remove impulse noise from color digital images. Nevertheless, there are a few works that have addressed impulse noise removal for grayscale images, the most relevant being [38][39][40].…”

Application of Motion Sensor Based on Neural Network in Basketball Technology and Physical Fitness Evaluation System

“…It can be seen from Figure 1 that the basic steps are as follows: wavelet transforms the signal, uses the threshold function to process each layer of wavelet separately, and gets the signal after wavelet reconstruction to remove noise. In the denoising process, by using the threshold function, the noise 3 Wireless Communications and Mobile Computing below the threshold can be effectively filtered out [20]. If the selected threshold is too large, part of the effective information of the signal will be filtered out, resulting in serious signal distortion; if the selected threshold is too small, the denoising will be incomplete.…”