Contrastive adaptive frequency decomposition network guided by haze discrimination for real-world image dehazing

薄雾 计算机科学 分解 人工智能 图像(数学) 计算机视觉 物理 生物 生态学 气象学
作者
Yaozong Mo,Chaofeng Li
出处
期刊:Displays [Elsevier BV]
卷期号:82: 102665-102665 被引量:1
标识
DOI:10.1016/j.displa.2024.102665
摘要

Recent unsupervised image dehazing methods used unpaired real-world training data for enhancing generalization on real-world scenes. However, these methods often require dehazing and rehazing cycles with auxiliary networks for training, resulting in high computational costs and extended training time. In this work, we propose an unsupervised dehazing framework called Contrastive Adaptive Frequency Decomposition Dehazing Network (CAFDD). By incorporating carefully designed network structure and constraints, our CAFDD well avoids additional training overhead and needs only 1.91M parameters. Specifically, we first consider the following insights, including: 1) Haze primarily affects high-frequency components in an image, resulting in blurred edges; 2) Low-frequency components capture the large-scale variations with less susceptibility to haze; and 3) Existing unlearnable frequency decomposition methods such Fourier transform often suffer from information loss, and thus develop the novel PMP (Pointwise convolution-Max pooling-Pointwise convolution) and DAD (Depthwise convolution-Average pooling-Depthwise convolution) blocks to automatically extract high and low-frequency features from input images for accurately estimating transmission map. Then, we propose haze discrimination (HD), a new pretext task for contrastive learning in image dehazing, by forming positive and negative pairs based on haze presence, in order for guiding the network to extract visibility-related features. Last, to get rid of the rehazing cycle and improve training efficiency, we construct a pixel-level constraint, histogram equalization-based texture loss function, which enhances the sharpness and realism of the generated images. Through extensive experiments, we demonstrate the superiority of our CAFDD over the state-of-the-art dehazing approaches on real-world land and overwater images.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
刚刚
1sss完成签到,获得积分10
1秒前
1秒前
传奇3的应助被zhaoxiao采纳,获得10
1秒前
心灵美的幼蓉完成签到,获得积分10
1秒前
11完成签到 ,获得积分10
1秒前
辞旧完成签到,获得积分10
1秒前
2秒前
yjj发布了新的文献求助10
3秒前
坚定夜蕾完成签到,获得积分10
3秒前
Alphaz9918完成签到,获得积分10
3秒前
3秒前
假装忧郁被发现完成签到 ,获得积分10
3秒前
4秒前
zetero发布了新的文献求助10
4秒前
毕竟话少完成签到,获得积分10
5秒前
鲤鱼盼夏发布了新的文献求助10
5秒前
彭于晏的应助被零一采纳,获得10
6秒前
7秒前
7秒前
Hello的应助被cd采纳,获得10
7秒前
7秒前
penghui发布了新的文献求助10
8秒前
Wendy完成签到,获得积分10
8秒前
er发布了新的文献求助10
9秒前
SciGPT的应助被Summer夏天采纳,获得10
10秒前
10秒前
活泼的煎饼完成签到,获得积分10
10秒前
DOC_XIONG的应助被lc采纳,获得10
10秒前
清白之年的应助被lc采纳,获得10
11秒前
秒速五厘米的应助被lc采纳,获得10
11秒前
秒速五厘米的应助被lc采纳,获得10
11秒前
11秒前
Yyy完成签到 ,获得积分10
11秒前
12秒前
12秒前
zhaoxiao发布了新的文献求助10
12秒前
13秒前
再沉默完成签到,获得积分10
13秒前
瑜格发布了新的文献求助10
13秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1500
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
八旗制度の研究 600
Analyses de reconnaissances de dettes passées devant les échevins d'Ypres (1249-1291) 600
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8161158
求助须知:如何正确求助?哪些是违规求助? 9555911
关于积分的说明 21731165
捐赠科研通 7751082
什么是DOI,文献DOI怎么找? 3427515
关于科研通互助平台的介绍 2552547
邀请新用户注册赠送积分活动 2471950