Research on Defogging Algorithm Based on DRSformer

DOI

10.2991/978-94-6463-512-6_51How to use a DOI?

Keywords

Image Dehazing; Transformer Networks; Atmospheric Correction

Abstract

Recently, advancements in remote sensing technology have led to the collection of a large volume of remote sensing images. Nevertheless, these images are always influenced by atmospheric phenomena such as haze, which reduce their clarity and quality. Traditional dehazing methods and convolutional neural network (CNN)-based techniques show certain limitations when dealing with the complex and uneven pattern of haze in remote sensing images. The article repurposes DRSformer, a pioneering transformer network created for rain removal tasks, to tackle the problem of image dehazing in this research. DRSformer utilizes Sparse Transformer Blocks (STB) and a Mixture of Experts Feature Compensator (MEFC) to effectively address the challenges posed by nonuniform haze scenarios. Based on experimental results, DRSformer achieved good performance. It surpasses current appeached, achieving superior PSNR and SSIM values under various haze conditions. Furthermore, qualitative assessments indicate that DRSformer significantly improves visual clarity and detail preservation. Looking ahead, the adaptability of DRSformer can be further explored to enhance its performance under other atmospheric conditions and expand its applicability to a wider range of remote sensing image processing tasks.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Peizhou Huang
PY  - 2024
DA  - 2024/09/23
TI  - Research on Defogging Algorithm Based on DRSformer
BT  - Proceedings of the 2024 International Conference on Artificial Intelligence and Communication (ICAIC 2024)
PB  - Atlantis Press
SP  - 477
EP  - 487
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-512-6_51
DO  - 10.2991/978-94-6463-512-6_51
ID  - Huang2024
ER  -