DFR-Net: Density Feature Refinement Network for Image Dehazing Utilizing Haze Density Difference

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Zhongze, Zhao, Haitao, Yao, Lujian, Peng, Jingchao, Zhao, Kaijie
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913332123926528
author Wang, Zhongze
Zhao, Haitao
Yao, Lujian
Peng, Jingchao
Zhao, Kaijie
author_facet Wang, Zhongze
Zhao, Haitao
Yao, Lujian
Peng, Jingchao
Zhao, Kaijie
contents In image dehazing task, haze density is a key feature and affects the performance of dehazing methods. However, some of the existing methods lack a comparative image to measure densities, and others create intermediate results but lack the exploitation of their density differences, which can facilitate perception of density. To address these deficiencies, we propose a density-aware dehazing method named Density Feature Refinement Network (DFR-Net) that extracts haze density features from density differences and leverages density differences to refine density features. In DFR-Net, we first generate a proposal image that has lower overall density than the hazy input, bringing in global density differences. Additionally, the dehazing residual of the proposal image reflects the level of dehazing performance and provides local density differences that indicate localized hard dehazing or high density areas. Subsequently, we introduce a Global Branch (GB) and a Local Branch (LB) to achieve density-awareness. In GB, we use Siamese networks for feature extraction of hazy inputs and proposal images, and we propose a Global Density Feature Refinement (GDFR) module that can refine features by pushing features with different global densities further away. In LB, we explore local density features from the dehazing residuals between hazy inputs and proposal images and introduce an Intermediate Dehazing Residual Feedforward (IDRF) module to update local features and pull them closer to clear image features. Sufficient experiments demonstrate that the proposed method achieves results beyond the state-of-the-art methods on various datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2307_13927
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle DFR-Net: Density Feature Refinement Network for Image Dehazing Utilizing Haze Density Difference
Wang, Zhongze
Zhao, Haitao
Yao, Lujian
Peng, Jingchao
Zhao, Kaijie
Computer Vision and Pattern Recognition
In image dehazing task, haze density is a key feature and affects the performance of dehazing methods. However, some of the existing methods lack a comparative image to measure densities, and others create intermediate results but lack the exploitation of their density differences, which can facilitate perception of density. To address these deficiencies, we propose a density-aware dehazing method named Density Feature Refinement Network (DFR-Net) that extracts haze density features from density differences and leverages density differences to refine density features. In DFR-Net, we first generate a proposal image that has lower overall density than the hazy input, bringing in global density differences. Additionally, the dehazing residual of the proposal image reflects the level of dehazing performance and provides local density differences that indicate localized hard dehazing or high density areas. Subsequently, we introduce a Global Branch (GB) and a Local Branch (LB) to achieve density-awareness. In GB, we use Siamese networks for feature extraction of hazy inputs and proposal images, and we propose a Global Density Feature Refinement (GDFR) module that can refine features by pushing features with different global densities further away. In LB, we explore local density features from the dehazing residuals between hazy inputs and proposal images and introduce an Intermediate Dehazing Residual Feedforward (IDRF) module to update local features and pull them closer to clear image features. Sufficient experiments demonstrate that the proposed method achieves results beyond the state-of-the-art methods on various datasets.
title DFR-Net: Density Feature Refinement Network for Image Dehazing Utilizing Haze Density Difference
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2307.13927