UIR-Net: A Simple and Effective Baseline for Underwater Image Restoration and Enhancement

Long²,

ACM Trans. Multimedia Comput. Commun. Appl.

et al. 2023

Residual image and illumination estimation have been proven to be helpful for image enhancement. In this paper, we propose a general framework, called RI-GAN, that exploits residual and illumination using generative adversarial networks (GANs). The proposed framework detects and removes shadows in a coarse-to-fine fashion. At the coarse stage, we employ three generators to produce a coarse shadow-removal result, a residual image, and an inverse illumination map. We also incorporate two indirect shadow-removal images via the residual image and the inverse illumination map. With the residual image, the illumination map, and the two indirect shadow-removal images as auxiliary information, the refinement stage estimates a shadow mask to identify shadow regions in the image, and then refines the coarse shadow-removal result to the fine shadow-free image. We introduce a cross-encoding module to the refinement generator, in which the use of feature-crossing can provide additional details to promote the shadow mask and the high-quality shadow-removal result. In addition, we apply data augmentation to the discriminator to reduce the dependence between representations of the discriminator and the quality of the predicted image. Experiments for shadow detection and shadow removal demonstrate that our method outperforms state-of-the-art methods. Furthermore, RI-GAN exhibits good performance in terms of image dehazing, rain removal, and highlight removal, demonstrating the effectiveness and flexibility of the proposed framework.

“…During the above iteration process, equation (14) has two unresolved problems. The first problem is to calculate the gradient of the objective function.…”

Section: B Variational Model Of Iterative Algorithm Cnn Frameworkmentioning

confidence: 99%

Exploiting Residual and Illumination with GANs for Shadow Detection and Shadow Removal

Long²,

ACM Trans. Multimedia Comput. Commun. Appl.

et al. 2023

“…A network of deep neural networks was implemented by Li et al [21] to descatter the underwater image. The use of a deep convolutional neural network (CNN) has been recommended by Perez et al [22] to dehaze underwater shots. Deep CNN was implemented by Wang et al [23] to color-correct and eliminate haze from underwater photographs.…”

Section: Related Workmentioning

confidence: 99%

A Hybrid Approach for Underwater Image Enhancement using CNN and GAN

Menon¹,

Aarthi²

2023

IJACSA

Underwater image-capturing technology has advanced over the years, and varieties of artificial intelligencebased applications have been developed on digital and synthetic images. The low-quality and low-resolution underwater images are challenging factors for use in existing image processing in computer vision applications. Degraded or low-quality photos are common issues in the underwater imaging process due to natural factors like low illumination and scattering. The recent techniques use deep learning architectures like CNN, GAN, or other models for image enhancement. Although adversarialbased architectures provide good perceptual quality, they performed worse in quantitative tests compared with convolutional-based networks. A hybrid technique is proposed in this paper that blends both designs to gain advantages of the CNN and GAN architectures. The generator component produces or makes images, which contributes to the creation of a sizable training set. The EUVP dataset is used for experimentation for model training and testing. The PSNR score was observed to measure the visual quality of the resultant images produced by models. The proposed system was able to provide an improved image with a higher PSNR score and SSIM score with state-of-the-art methods.

Hybrid regularization inspired by total variation and deep denoiser prior for image restoration

Liang,

Zhang,

Wei

et al. 2024

Complex Intell. Syst.

Image restoration is a fundamental problem in computer vision, with the goal of restoring high-quality images from degraded low-quality observation images. However, the ill-posedness of restoration problem often leads to artifacts in the results. It inspires us to study the combination of prior, effectively restrict the solution space and improving the quality of the restored image. In this paper, a novel hybrid regularization method for image restoration is proposed, which adopts both internal and external image priors into a unified framework. Specifically, a new hybrid regularization model is designed. The TV model and a deep image denoiser are inserted into the restoration model using the plug-and-play framework, protecting image details while implicitly using external information for deep denoising. Moreover, in order to make the proposed hybrid regularization operable, an adaptive parameter method is proposed to adaptively balance the TV model and learned model in iteration. Experiments show that the proposed method surpasses the performance of existing image restoration techniques.