Deep Dehazing Network for Remote Sensing Image with Non-Uniform Haze

Abstract

The haze in remote sensing images can cause the decline of image quality and bring many obstacles to the applications of remote sensing images. Considering the non-uniform distribution of haze in remote sensing images, we propose a single remote sensing image dehazing method based on the encoder–decoder architecture, which combines both wavelet transform and deep learning technology. To address the clarity issue of remote sensing images with non-uniform haze, we preliminary process the input image by the dehazing method based on the atmospheric scattering model, and extract the first-order low-frequency sub-band information of its 2D stationary wavelet transform as an additional channel. Meanwhile, we establish a large-scale hazy remote sensing image dataset to train and test the proposed method. Extensive experiments show that the proposed method obtains greater advantages over typical traditional methods and deep learning methods qualitatively. For the quantitative aspects, we take the average of four typical deep learning methods with superior performance as a comparison object using 500 random test images, and the peak-signal-to-noise ratio (PSNR) value using the proposed method is improved by 3.5029 dB, and the structural similarity (SSIM) value is improved by 0.0295, respectively. Based on the above, the effectiveness of the proposed method for the problem of remote sensing non-uniform dehazing is verified comprehensively.

DSSCNet: Deep Custom Spatial and Spectral Consistency Layer-Based Dehazing Network

“…These techniques have demonstrated superior performance compared to traditional methods, as they can capture intricate image features and implicitly learn the complex scattering and absorption phenomena. One of the key advantages of deep learning-based dehazing is its ability to generalize well to real-world scenarios, including those with non-uniform haze distributions, varying scene complexities, and diverse lighting conditions [6]. These methods can effectively restore details in challenging scenarios, such as underwater and nighttime dehazing, where traditional approaches often struggle to provide satisfactory results [7].…”

“…In this ever-evolving landscape of image dehazing, researchers strive to address the challenges posed by various environmental conditions and to achieve natural, artifact-free, and visually pleasing dehazed images [5]. The fusion of physical models, deep learning, and domain-specific information holds tremendous promise for pushing the boundaries of image dehazing and enabling practical and effective solutions for a wide range of realworld applications [6].…”

Voids Filling of DEM with Multiattention Generative Adversarial Network Model

“…where y i and y i represent the elevation value of each pixel of the filling result and the original data, and N is the total number of pixels in the void area of each datum. In addition, this paper uses peak signal-to-noise ratio (PSNR) [53] and structural similarity (SSIM) [54] for evaluation of the similarity structure of terrain surrounding the void.…”

Swin Transformer UNet for Very High Resolution Image Dehazing

“…Common window-based multi-head self-attention and shifted window multi-head self-attention are also employed in the patches. (9)…”

“…To train the proposed ST-UNet for the dehazing task, it is necessary to generate a dataset. We generate hazed images using a non-uniform haze-adding algorithm, (9) and we employ two typical aerial view and haze-free datasets as basic datasets for training, i.e., AID30 (10) and RSSCN8. (11) AID30 is a remote sensing dataset open sourced by Wuhan University, which is used for image recognition.…”