Zero-Shot Remote Sensing Image Dehazing Based on a Re-Degradation Haze Imaging Model
Image dehazing is crucial for improving the advanced applications on remote sensing (RS) images. However, collecting paired RS images to train the deep neural networks (DNNs) is scarcely available, and the synthetic datasets may suffer from domain-shift issues. In this paper, we propose a zero-shot...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-11-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/14/22/5737 |
_version_ | 1797464108898975744 |
---|---|
author | Jianchong Wei Yi Wu Liang Chen Kunping Yang Renbao Lian |
author_facet | Jianchong Wei Yi Wu Liang Chen Kunping Yang Renbao Lian |
author_sort | Jianchong Wei |
collection | DOAJ |
description | Image dehazing is crucial for improving the advanced applications on remote sensing (RS) images. However, collecting paired RS images to train the deep neural networks (DNNs) is scarcely available, and the synthetic datasets may suffer from domain-shift issues. In this paper, we propose a zero-shot RS image dehazing method based on a re-degradation haze imaging model, which directly restores the haze-free image from a single hazy image. Based on layer disentanglement, we design a dehazing framework consisting of three joint sub-modules to disentangle the hazy input image into three components: the atmospheric light, the transmission map, and the recovered haze-free image. We then generate a re-degraded hazy image by mixing up the hazy input image and the recovered haze-free image. By the proposed re-degradation haze imaging model, we theoretically demonstrate that the hazy input and the re-degraded hazy image follow a similar haze imaging model. This finding helps us to train the dehazing network in a zero-shot manner. The dehazing network is optimized to generate outputs that satisfy the relationship between the hazy input image and the re-degraded hazy image in the re-degradation haze imaging model. Therefore, given a hazy RS image, the dehazing network directly infers the haze-free image by minimizing a specific loss function. Using uniform hazy datasets, non-uniform hazy datasets, and real-world hazy images, we conducted comprehensive experiments to show that our method outperforms many state-of-the-art (SOTA) methods in processing uniform or slight/moderate non-uniform RS hazy images. In addition, evaluation on a high-level vision task (RS image road extraction) further demonstrates the effectiveness and promising performance of the proposed zero-shot dehazing method. |
first_indexed | 2024-03-09T18:02:17Z |
format | Article |
id | doaj.art-b1c2b17027314625b5c66130606ab4e7 |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-09T18:02:17Z |
publishDate | 2022-11-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-b1c2b17027314625b5c66130606ab4e72023-11-24T09:49:38ZengMDPI AGRemote Sensing2072-42922022-11-011422573710.3390/rs14225737Zero-Shot Remote Sensing Image Dehazing Based on a Re-Degradation Haze Imaging ModelJianchong Wei0Yi Wu1Liang Chen2Kunping Yang3Renbao Lian4College of Photonic and Electronic Engineering, Fujian Normal University, Fuzhou 350117, ChinaCollege of Photonic and Electronic Engineering, Fujian Normal University, Fuzhou 350117, ChinaCollege of Photonic and Electronic Engineering, Fujian Normal University, Fuzhou 350117, ChinaCollege of Photonic and Electronic Engineering, Fujian Normal University, Fuzhou 350117, ChinaCollege of Electronics and Information Science, Fujian Jiangxia University, Fuzhou 350108, ChinaImage dehazing is crucial for improving the advanced applications on remote sensing (RS) images. However, collecting paired RS images to train the deep neural networks (DNNs) is scarcely available, and the synthetic datasets may suffer from domain-shift issues. In this paper, we propose a zero-shot RS image dehazing method based on a re-degradation haze imaging model, which directly restores the haze-free image from a single hazy image. Based on layer disentanglement, we design a dehazing framework consisting of three joint sub-modules to disentangle the hazy input image into three components: the atmospheric light, the transmission map, and the recovered haze-free image. We then generate a re-degraded hazy image by mixing up the hazy input image and the recovered haze-free image. By the proposed re-degradation haze imaging model, we theoretically demonstrate that the hazy input and the re-degraded hazy image follow a similar haze imaging model. This finding helps us to train the dehazing network in a zero-shot manner. The dehazing network is optimized to generate outputs that satisfy the relationship between the hazy input image and the re-degraded hazy image in the re-degradation haze imaging model. Therefore, given a hazy RS image, the dehazing network directly infers the haze-free image by minimizing a specific loss function. Using uniform hazy datasets, non-uniform hazy datasets, and real-world hazy images, we conducted comprehensive experiments to show that our method outperforms many state-of-the-art (SOTA) methods in processing uniform or slight/moderate non-uniform RS hazy images. In addition, evaluation on a high-level vision task (RS image road extraction) further demonstrates the effectiveness and promising performance of the proposed zero-shot dehazing method.https://www.mdpi.com/2072-4292/14/22/5737remote sensing imageconvolutional neural networkzero-shot learningimage dehazing |
spellingShingle | Jianchong Wei Yi Wu Liang Chen Kunping Yang Renbao Lian Zero-Shot Remote Sensing Image Dehazing Based on a Re-Degradation Haze Imaging Model Remote Sensing remote sensing image convolutional neural network zero-shot learning image dehazing |
title | Zero-Shot Remote Sensing Image Dehazing Based on a Re-Degradation Haze Imaging Model |
title_full | Zero-Shot Remote Sensing Image Dehazing Based on a Re-Degradation Haze Imaging Model |
title_fullStr | Zero-Shot Remote Sensing Image Dehazing Based on a Re-Degradation Haze Imaging Model |
title_full_unstemmed | Zero-Shot Remote Sensing Image Dehazing Based on a Re-Degradation Haze Imaging Model |
title_short | Zero-Shot Remote Sensing Image Dehazing Based on a Re-Degradation Haze Imaging Model |
title_sort | zero shot remote sensing image dehazing based on a re degradation haze imaging model |
topic | remote sensing image convolutional neural network zero-shot learning image dehazing |
url | https://www.mdpi.com/2072-4292/14/22/5737 |
work_keys_str_mv | AT jianchongwei zeroshotremotesensingimagedehazingbasedonaredegradationhazeimagingmodel AT yiwu zeroshotremotesensingimagedehazingbasedonaredegradationhazeimagingmodel AT liangchen zeroshotremotesensingimagedehazingbasedonaredegradationhazeimagingmodel AT kunpingyang zeroshotremotesensingimagedehazingbasedonaredegradationhazeimagingmodel AT renbaolian zeroshotremotesensingimagedehazingbasedonaredegradationhazeimagingmodel |