Abstract:Space probes are always obstructed by floating objects in the atmosphere (clouds, haze, rain, etc.) during imaging, resulting in the loss of a significant amount of detailed information in remote sensing images and severely reducing the quality of the remote sensing images. To address the problem of detailed information loss in remote sensing images, we propose an end-to-end detail enhancement network to directly remove haze in remote sensing images, restore detailed information of the image, and improve the quality of the image. In order to enhance the detailed information of the image, we designed a multi-scale detail enhancement unit and a stepped attention detail enhancement unit, respectively. The former extracts multi-scale information from images, integrates global and local information, and constrains the haze to enhance the image details. The latter uses the attention mechanism to adaptively process the uneven haze distribution in remote sensing images from three dimensions: deep, middle and shallow. It focuses on effective information such as haze and high frequency to further enhance the detailed information of the image. In addition, we embed the designed parallel normalization module in the network to further improve the dehazing performance and robustness of the network. Experimental results on the SateHaze1k and HRSD datasets demonstrate that our method effectively handles remote sensing images obscured by various levels of haze, restores the detailed information of the images, and outperforms the current state-of-the-art haze removal methods.

What problem does this paper attempt to address?

### Problems the paper attempts to solve The paper aims to solve the problem of loss of detailed information in remote - sensing images. Specifically, when a space probe performs imaging through the atmosphere, floating objects in the atmosphere (such as clouds, haze, rain, etc.) will interfere with the imaging process, resulting in the loss of a large amount of detailed information in remote - sensing images and severely degrading the image quality. To solve this problem, the author proposes an end - to - end Detail - Enhancement Dehazing Network (EDED - Net), which directly removes haze from remote - sensing images, restores the detailed information of the images, and improves the image quality. ### Main contributions 1. **Multi - Scale Detail - Enhancement Unit (MSDEU)**: - It can extract multi - scale features of the input image and integrate global and local information. - Using parallel dilated convolutions with a large receptive field and long - range modeling ability, enabling the network to capture context information across a wide range of spatial scales and effectively enhance image details. 2. **Step - by - Step Attention Detail - Enhancement Unit (SADEU)**: - It adaptively focuses on the high - frequency information of the image from three dimensions (deep, medium, and shallow). - It can flexibly handle unevenly distributed haze and is more suitable for removing haze in remote - sensing images. 3. **Parallel Normalization Module (PNM)**: - It can simultaneously learn features related to content and not affected by appearance changes, effectively improving the generalization ability and robustness of the network. 4. **End - to - End Detail - Enhancement Dehazing Network (EDED - Net)**: - It embeds MSDEU and SADEU into EDED - Net for image dehazing in remote - sensing scenes. - On challenging benchmark datasets (such as SateHaze1k and HRSD), this method outperforms the current state - of - the - art dehazing methods and can more effectively remove haze in remote - sensing images. ### Method overview - **Network architecture**: EDED - Net is an improved 5 - stage U - Net, in which the convolution blocks of U - Net are replaced by designed Detail - Enhancement Blocks (DEB). DEB is composed of MSDEU and SADEU and is used to adaptively extract multi - scale high - frequency, edge and other detailed information. - **Multi - Scale Detail - Enhancement Unit (MSDEU)**: - It uses parallel dilated convolutions to extract features at different scales. - It enhances high - frequency details and edge information through local residual connections. - **Step - by - Step Attention Detail - Enhancement Unit (SADEU)**: - It uses Channel Attention (CA) to extract globally shared information. - It uses Pixel Attention (PA) to extract position - related local information. - It designs Deep Pixel Attention (DPA) and Shallow Pixel Attention (SPA) to focus on different levels of haze and high - frequency information respectively. - **Parallel Normalization Module (PNM)**: - It combines instance normalization and batch normalization to improve the generalization ability and robustness of the network. ### Experimental results The experimental results show that EDED - Net effectively processes remote - sensing images occluded by different degrees of haze on the SateHaze1k and HRSD datasets, restores the detailed information of the images, and outperforms the current state - of - the - art dehazing methods.

End-to-End Detail-Enhanced Dehazing Network for Remote Sensing Images

Nonuniformly Dehaze Network for Visible Remote Sensing Images

From Coarse to Fine: A Two-Stage Network for Dense Haze removal

Multi-Scale Attention Feature Enhancement Network for Single Image Dehazing

DehazeNet: An End-to-End System for Single Image Haze Removal

Single Image Dehazing Based on Two-Stream Convolutional Neural Network

Depth-Guided Dehazing Network for Long-Range Aerial Scenes

AIDEDNet: anti-interference and detail enhancement dehazing network for real-world scenes

Residual Spatial and Channel Attention Networks for Single Image Dehazing

A new end-to-end image dehazing algorithm based on residual attention mechanism

Single image mixed dehazing method based on numerical iterative model and DehazeNet

Atmospheric Light Estimation Based Remote Sensing Image Dehazing

A Coarse-to-Fine Two-Stage Attentive Network for Haze Removal of Remote Sensing Images

Lightweight single image dehazing network with residual feature attention

Remote sensing image dehazing based on data blending and Laplace network

MDEANet: modified detail-enhanced convolution and attention-based network for dehazing of remote sensing images

Haze-Aware Attention Network for Single-Image Dehazing

Hybrid High-Resolution Learning for Single Remote Sensing Satellite Image Dehazing

Haze Relevant Feature Attention Network for Single Image Dehazing

A Dehazing Method for Remote Sensing Image Under Nonuniform Hazy Weather Based on Deep Learning Network

Perceiving and Modeling Density is All You Need for Image Dehazing