双向映射与物理负相关的航空图像去雾算法

doi:10.7527/S1000-6893.2025.31652

Abstract

Abstract:

Recently， deep neural networks based aerial image dehazing techniques have achieved remarkable performance. However， most methods only learn a one-way mapping from hazy images to clear images， which fail to fully exploit the latent supervision information of haze generation， leading to dehazing performance degradation. Moreover， existing methods that incorporate the atmospheric scattering model in the feature space overlook the exploration of the negative correlation between the transmission map and atmospheric light， which is crucial for leaning more representative features. To address these issues， we propose a bidirectional mapping physics-aware negative correlation algorithm for aerial image dehazing. Specifically， the proposed bidirectional mapping multi-stage framework synchronously models haze removal and generation processes in a parameter-sharing network， effectively exploring the haze formation mechanism to enhance dehazing performance. Furthermore， an atmospheric light-guided physics-aware negative correlation module is designed， which optimizes transmission map estimation by leveraging a negative correlation guidance mechanism utilizing atmospheric light information. Experimental results on public datasets demonstrate that the proposed dehazing algorithm outperforms seven state-of-the art haze removal methods in both objective evaluation and visual quality.

Key words: aerial image dehazing, bidirectional mapping, multi-stage, negative correlation, atmospheric scattering model

CLC Number:

Zhaoru YAO, Hang SUN, Dong REN, Li LIU, Jun WAN. Bidirectional mapping and negative correlation algorithm for aerial image dehazing[J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(23): 631652.

Figures/Tables 9

Fig.1

Fig.2

Table 1

Table 2

Fig.3

Fig.4

Fig.5

Table 3

Fig. 6

References 39

[1]	SUN S L， WANG T Y， YANG H X， et al. An environmentally adaptive and contrastive representation learning method for condition monitoring of industrial assets［J］. IEEE Transactions on Cybernetics， 2024， 54（3）： 1484-1496.
[2]	SUNG T， KANG Y， IM J. Enhancing satellite-based wildfire monitoring： Advanced contextual model using environmental and structural information［J］. IEEE Transactions on Geoscience and Remote Sensing， 2024， 62： 4409516.
[3]	EVANGELISTA S S， MATURAN F， WENCESLAO C， et al. Cumulative prospect theory under different types of input data for public health resilience assessment during natural disasters［J］. Expert Systems with Applications， 2024， 258： 125172.
[4]	CHEN F， SUN Y， WANG L， et al. HRTBDA： A network for post-disaster building damage assessment based on remote sensing images［J］. International Journal of Digital Earth， 2024， 17（1）：2418880.
[5]	YAN L Q， YANG S H， ZHANG Q， et al. Multi-source information fusion attention network for weakly supervised salient object detection in optical remote sensing images［J］. Expert Systems with Applications， 2025， 261： 125505.
[6]	ZHU J Y， QIN X B， ELSADDIK A. DC-Net： Divide-and-conquer for salient object detection［J］. Pattern Recognition， 2025， 157： 110903.
[7]	李红光，于若男，丁文锐. 基于深度学习的小目标检测研究进展［J］. 航空学报， 2021， 42（7）： 024691.
	LI H G， YU R N， DING W R. Research development of small object traching based on deep learning［J］. Acta Aeronautica et Astronautica Sinica， 2021， 42（7）： 024691 （in Chinese）.
[8]	FATIMA R， SADIQ R， ULLAH I， et al. Multiple passive-sensor distributed target tracking approach with Machine Learning Feedback［J］. Expert Systems with Applications， 2024， 238： 122344.
[9]	XUE X R， WEI D Z， HUANG S C. Cluster target tracking based on multi-sensor continuous-discrete PMBM filter［J］. Expert Systems with Applications， 2024， 245： 123121.
[10]	LI W， LIAO M X， ZOU W B. A progressive segmentation network for navigable areas with semantic-spatial information flow［J］. Expert Systems with Applications， 2025， 261： 125465.
[11]	ZHENG X， LUO Y H， ZHOU P Y， et al. Distilling efficient vision transformers from CNNs for semantic segmentation［J］. Pattern Recognition， 2025， 158： 111029.
[12]	HE K M， SUN J， TANG X O. Single image haze removal using dark channel prior［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2011， 33（12）： 2341-2353.
[13]	ZHU Q S， MAI J M， SHAO L. A fast single image haze removal algorithm using color attenuation prior［J］. IEEE Transactions on Image Processing， 2015， 24（11）： 3522-3533.
[14]	BERMAN D， TREIBITZ T， AVIDAN S. Non-local image dehazing［C］∥2016 IEEE Conference on Computer Vision and Pattern Recognition （CVPR）. Piscataway： IEEE Press， 2016.
[15]	CANTOR A. Optics of the atmosphere： Scattering by molecules and particles［J］. IEEE Journal of Quantum Electronics， 1978， 14（9）： 698-699.
[16]	QIN X， WANG Z L， BAI Y C， et al. FFA-net： Feature fusion attention network for single image dehazing［J］. Proceedings of the AAAI Conference on Artificial Intelligence， 2020， 34（7）： 11908-11915.
[17]	DONG J X， PAN J S. Physics-based feature dehazing networks［C］∥Computer Vision-ECCV 2020. Cham： Springer International Publishing， 2020.
[18]	ZHENG Y， ZHAN J H， HE S F， et al. Curricular contrastive regularization for physics-aware single image dehazing［C］∥2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）. Piscataway： IEEE Press， 2023.
[19]	WEN Y B， GAO T， ZHANG J， et al. Encoder-free multiaxis physics-aware fusion network for remote sensing image dehazing［J］. IEEE Transactions on Geoscience and Remote Sensing， 2023， 61： 4705915.
[20]	SONG Y D， HE Z Q， QIAN H， et al. Vision transformers for single image dehazing［J］. IEEE Transactions on Image Processing， 2023， 32： 1927-1941.
[21]	CHEN Z X， HE Z W， LU Z M. DEA-net： Single image dehazing based on detail-enhanced convolution and content-guided attention［J］. IEEE Transactions on Image Processing， 2024， 33： 1002-1015.
[22]	CUI Y N， REN W Q， CAO X C， et al. Image restoration via frequency selection［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2023， 46（2）： 1093-1108.
[23]	CUI Y N， KNOLL A. Dual-domain strip attention for image restoration［J］. Neural Networks， 2024， 171： 429-439.
[24]	GUO Y， CHEN J， WANG J D， et al. Closed-loop matters： Dual regression networks for single image super-resolution［C］∥2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）. Piscataway： IEEE Press， 2020.
[25]	ZHU Y R， HUANG J， FU X Y， et al. Bijective mapping network for shadow removal［C］∥2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）. Piscataway： IEEE Press， 2022.
[26]	FENG C F， CHEN Z Y， KOU R K， et al. HazyDet： Open-source benchmark for drone-view object detection with depth-cues in hazy scenes［DB/OL］. arXiv preprint： 2409.19833，2024.
[27]	ZHANG L B， WANG S. Dense haze removal based on dynamic collaborative inference learning for remote sensing images［J］. IEEE Transactions on Geoscience and Remote Sensing， 2022， 60： 5631016.
[28]	CAI B L， XU X M， JIA K， et al. DehazeNet： An end-to-end system for single image haze removal［J］. IEEE Transactions on Image Processing， 2016， 25（11）： 5187-5198.
[29]	REN W Q， LIU S， ZHANG H， et al. Single image dehazing via multi-scale convolutional neural networks［M］∥Computer Vision-ECCV 2016. Cham： Springer International Publishing， 2016： 154-169.
[30]	ZHANG H， PATEL V M. Densely connected pyramid dehazing network［C］∥2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE Press， 2018.
[31]	HE T， ZHANG Z， ZHANG H， et al. Bag of tricks for image classification with convolutional neural networks［C］∥2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）. Piscataway： IEEE Press， 2019.
[32]	WANG Z， BOVIK A C， SHEIKH H R， et al. Image quality assessment： From error visibility to structural similarity［J］. IEEE Transactions on Image Processing， 2004， 13（4）： 600-612.
[33]	VENKATANATH N， PRANEETH D， BH M C， et al. Blind image quality evaluation using perception based features［C］∥IEEE National Conference on Communications. Piscataway： IEEE Press， 2015.
[34]	MITTAL A， MOORTHY A K， BOVIK A C. Blind/referenceless image spatial quality evaluator［C］∥2011 Conference Record of the Forty Fifth Asilomar Conference on Signals， Systems and Computers （ASILOMAR）. Piscataway： IEEE Press， 2011.
[35]	MITTAL A， SOUNDARARAJAN R， BOVIK A C. Making a “completely blind” image quality analyzer［J］. IEEE Signal Processing Letters， 2013， 20（3）： 209-212.
[36]	王敬东，张文涛，王子瑞，等. 一种快速航空图像去雾算法［J］. 航空学报， 2013， 34（3）： 636-643.
	WANG J D， ZHANG W T， WANG Z R， et al. A fast aerial image de-haze algorithm［J］. Acta Aeronautica et Astronautica Sinica， 2013， 34（3）： 636-643 （in Chinese）.
[37]	孙航，方帅领，但志平，等. 层级特征交互与增强感受野双分支遥感图像去雾网络［J］. 遥感学报， 2023， 27（12）： 2831-2846.
	SUN H， FANG S L， DAN Z P， et al. A two-branch remote sensing image dehazing network based on hierarchical feature interaction and enhanced receptive field［J］. National Remote Sensing Bulletin， 2023， 27（12）： 2831-2846 （in Chinese）.
[38]	LI H Y， LI J， ZHAO D， et al. DehazeFlow： Multi-scale conditional flow network for single image dehazing［C］∥Proceedings of the 29th ACM International Conference on Multimedia. New York： ACM， 2021.
[39]	GUO X J， DU Y Q， ZHAO L. Property controllable variational autoencoder via invertible mutual dependence ［C］∥International Conference on Learning Representations， 2021. Vienna： OpenReview， 2021.

方法	文献	年份	HazyDet			DHID	Overhead
方法	文献	年份	PSNR	SSIM	PSNR	SSIM	Parameters/10⁶	FLOPs/10⁹	Runtime/ms	Weight Size/MB
DCP	［12］	2010	16.98	0.824	11.37	0.621		0.6	10.48
FFA	［16］	2020	19.25	0.798	22.03	0.840	4.69	287.53	45.82	17.88
C2PNet	［18］	2023	21.31	0.832	27.35	0.881	7.17	460.95	80.60	27.34
DehazeFormer	［20］	2023	17.53	0.802	27.23	0.887	2.51	25.79	45.27	9.57
DEA-Net	［21］	2024	25.01	0.890	27.42	0.891	13.65	32.23	32.32	52.05
FSNet	［22］	2024	24.14	0.884	27.46	0.897	13.28	111.14	59.89	50.64
DSANet	［23］	2024	19.01	0.751	26.84	0.880	3.86	37.72	16.14	14.72
本文方法			27.06	0.908	27.58	0.899	1.01	29.45	24.01	3.85

方法	PIQE	BRISQUE	NIQE
Hazy image	28.57	40.51	4.857
DCP	40.19	32.34	4.352
FFA	23.66	28.76	4.512
C2PNet	23.55	24.65	4.654
DehazeFormer	27.77	29.13	4.478
DEA-Net	19.11	23.57	4.406
FSNet	21.31	24.02	4.464
DSANet	25.34	28.55	4.754
本文方法	16.51	23.39	4.313

方法	PSNR	SSIM	Parameters/10⁶	FLOPs/10⁹
Base	23.81	0.861	1.99	57.29
Base+创新1	26.33	0.894	1.01	29.45
Base+创新2	26.20	0.892	1.24	29.68
本文方法	27.06	0.908	1.24	29.68

Bidirectional mapping and negative correlation algorithm for aerial image dehazing

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 9

References 39

Related Articles 9

Recommended Articles

Metrics

Comments

[1]	Lili CHEN, Jianxia LIU, Juntao ZHANG, Zheng GUO, Anping WU, Zhongxi HOU. Waverider forebody design method with longitudinal segments and multi-stage compression [J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(4): 128744-128744.
[2]	Cancan LI, Zuoli XIAO, Jiaqi LUO. Impact of pressure variations at inlet and outlet boundaries on aerodynamic performance of multi-stage axial compressor [J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(19): 630168-630168.
[3]	Ronghui CHENG, Huawei YU, Songbai WANG, Lin DU, Dakun SUN, Xiaofeng SUN. Effect of aerodynamic loading coefficient on occurrence of non-synchronous vibration in a multi-stage compressor [J]. Acta Aeronautica et Astronautica Sinica, 2023, 44(14): 628722-628722.
[4]	LUO Jiaqi, YANG Jing. Aerodynamic design optimization of a single low-speed compressor stage by an adjoint method [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2020, 41(5): 623368-623368.
[5]	YANG Xiaofei, JIANG Yongsong, PAN Ruochi, MENG Dejun. 2D performance prediction approach for multi-stage axial compressors [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2020, 41(3): 123515-123515.
[6]	ZHAO Huan, JIAO Zhongze, SUN Dan, LIU Yongquan, ZHAN Peng, XIN Qi. Numerical and experimental research on interstage pressure drop distribution affecting factors of multi-stage brush seals [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2020, 41(10): 123544-123544.
[7]	LIN Xiaojun, CUI Tong, YANG Biying, YANG Rui, XIN Xiaopeng. Method for establishing machining and inspection model of multi-stage machining processes of thin-walled blades [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2019, 40(11): 423034-423034.
[8]	DAI Zhengliang, BA Bin, ZHANG Yankui, CUI Weijia, WANG Daming. Fast DOA estimation for coherently distributed noncircular sources by backtracking reduced dimension [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2017, 38(9): 321034-321034.
[9]	Zhang Huisheng;Lin Chongping. A MICROPROCESSOR-BASED AUTOMATION TEST SYSTEM FOR THE EXPERIMENT OF THE MULTI-STAGE COMPRESSOR [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 1991, 12(8): 447-450.