Image Denoising Using a Novel Deep Generative Network with Multiple Target Images and Adaptive Termination Condition

Abstract

Image denoising, a classic ill-posed problem, aims to recover a latent image from a noisy measurement. Over the past few decades, a considerable number of denoising methods have been studied extensively. Among these methods, supervised deep convolutional networks have garnered increasing attention, and their superior performance is attributed to their capability to learn realistic image priors from a large amount of paired noisy and clean images. However, if the image to be denoised is significantly different from the training images, it could lead to inferior results, and the networks may even produce hallucinations by using inappropriate image priors to handle an unseen noisy image. Recently, deep image prior (DIP) was proposed, and it overcame this drawback to some extent. The structure of the DIP generator network is capable of capturing the low-level statistics of a natural image using an unsupervised method with no training images other than the image itself. Compared with a supervised denoising model, the unsupervised DIP is more flexible when processing image content that must be denoised. Nevertheless, the denoising performance of DIP is usually inferior to the current supervised learning-based methods using deep convolutional networks, and it is susceptible to the over-fitting problem. To solve these problems, we propose a novel deep generative network with multiple target images and an adaptive termination condition. Specifically, we utilized mainstream denoising methods to generate two clear target images to be used with the original noisy image, enabling better guidance during the convergence process and improving the convergence speed. Moreover, we adopted the noise level estimation (NLE) technique to set a more reasonable adaptive termination condition, which can effectively solve the problem of over-fitting. Extensive experiments demonstrated that, according to the denoising results, the proposed approach significantly outperforms the original DIP method in tests on different databases. Specifically, the average peak signal-to-noise ratio (PSNR) performance of our proposed method on four databases at different noise levels is increased by 1.90 to 4.86 dB compared to the original DIP method. Moreover, our method achieves superior performance against state-of-the-art methods in terms of popular metrics, which include the structural similarity index (SSIM) and feature similarity index measurement (FSIM). Thus, the proposed method lays a good foundation for subsequent image processing tasks, such as target detection and super-resolution.

Signal-to-Noise Ratio Comparison of Several Filters against Phantom Image

“…However, wavelet transform has one of the common disadvantages such as poor directionality and shift invariance [ 18 ]. Similarly, in [ 19 ], the authors proposed a new deep generative network coupled with several target images and adaptive termination situation. They generated two clear target images by using a normal denoising technique that enabled best guidance during conjunction step and improved its speed.…”

Image Denoising Using a Compressive Sensing Approach Based on Regularization Constraints

“…Many inverse problem optimization approaches for image denoising have been proposed in the literature. Some are based on deep learning and, more precisely, on the deep generative network [ 16 ], and others are based on models [ 17 ].…”

Denoising Single Images by Feature Ensemble Revisited

“…Liu et al [7] introduced GradNet by revisiting the image gradient theory of neural networks. Recently, several GAN-based approaches [29][30][31][32] were introduced through generating denoised images following either a data augmentation strategy for creating diverse training samples or a strategy based on the distribution of the clean images.…”