Efficient Image Dehazing with Boundary Constraint and Contextual Regularization

By Gaofeng Meng, Ying Wang, Jiangyong Duan, Shiming Xiang, Chunhong Pan. IEEE International Conference on Computer Vision, 2013

1,081 citations recorded by the source corpus.

Images captured in foggy weather conditions often suffer from bad visibility. In this paper, we propose an efficient regularization method to remove hazes from a single input image. Our method benefits much from an exploration on the inherent boundary constraint on the transmission function. This constraint, combined with a weighted L1-norm based contextual regularization, is modeled into an optimization problem to estimate the unknown scene transmission. A quite efficient algorithm based on variable splitting is also presented to solve the problem. The proposed method requires only a few general assumptions and can restore a high-quality haze-free image with faithful colors and fine image details. Experimental results on a variety of haze images demonstrate the effectiveness and efficiency of the proposed method.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/e2b99083f3b88691af5553b2a715c1fe43e59246

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Rain Streak Removal Using Layer Priors (2016, neutral): “[17] as preprocessing is useful.”
Dense-Haze: A Benchmark for Image Dehazing with Dense-Haze and Haze-Free Images (2019, background): “2, we can observe that in general the DCP-based techniques [5], [15], [35] recover the global image structure.”
Benchmarking Single-Image Dehazing and Beyond (2017, methodology): “DCP [9] FVR [10] BCCR [11] GRM [12] CAP [13] NLD [14] DehazeNet [15] MSCNN [16] AOD-Net [17] PSNR 16.”
AOD-Net: All-in-One Dehazing Network (2017, methodology): “We compared the proposed model with several stateof-the-art dehazing methods: Fast Visibility Restoration (FVR) [25], Dark-Channel Prior (DCP) [8], Boundary Constrained Context Regularization (BCCR) [12], Automatic Atmospheric Light Recovery (ATM) [22], Color Attenuation Prior (CAP) [32], Non-local ”
Domain Adaptation for Image Dehazing (2020, methodology): “Follow-up works have improved the efficiency and performance of the DCP method [30, 22, 17, 24, 34].”
Real-World Underwater Enhancement: Challenges, Benchmarks, and Solutions Under Natural Light (2019, methodology): “We applied our RUIE benchmark to quantitatively and qualitatively evaluate eleven representative underwater image enhancement algorithms 4 including Multi-Scale Retinex with Color Restore (MSRCR) [1], Contrast Limited Adaptive Histogram Equalization (CLAHE) [2], Fusion [3], Bianco Prior (BP) [5], Un”
You Only Look Yourself: Unsupervised and Untrained Single Image Dehazing Neural Network (2020, methodology): “, 2009), FVR (Tarel and Hautiere, 2009), BCCR (Meng et al., 2013), GRM (Chen et al.”
“Double-DIP”: Unsupervised Image Decomposition via Coupled Deep-Image-Priors (2018, background): “[23, 28, 18, 6, 8]) assume a uniform airlight color A for the entire image (i.”
DehazeNet: An End-to-End System for Single Image Haze Removal (2016, methodology): “To verify the effectiveness on complete images, DehazeNet is tested on synthesized hazy images from stereo images with a known depth map d (x), and it is compared with DCP [9], FVR [38], BCCR [11], ATM [39], CAP (2)[18], and RF [17].”
Gated Fusion Network for Single Image Dehazing (2018, methodology): “The second is Boundary Constrained Context Regularization (BCCR) proposed by Meng et al .”
Gated Context Aggregation Network for Image Dehazing and Deraining (2018, methodology): “To compensate for the lost information during the corruption procedure, many traditional methods [2, 16, 17, 29, 30, 46] leverage some image priors and visual cues to estimate the transmission maps and atmospheric light.”
Single Image Dehazing via Multi-scale Convolutional Neural Networks with Holistic Edges (2019, methodology): “Berman et al. (2016) introduce a non-local method for single image dehazing. based on the assumption that an image can be faithfully represented with just a few hundreds of distinct colors.”
I-HAZE: a dehazing benchmark with real hazy and haze-free indoor images (2018, background): “[18] approach, built also on dark channel prior, as expected, generates similar results as He et al.”
LIME: Low-Light Image Enhancement via Illumination Map Estimation (2017, methodology): “As mentioned, another widely used model is based on the observation that inverted low-light images 1−L look similar to haze images, which is thus expressed as [20], [21], [22]:”
Cycle-Dehaze: Enhanced CycleGAN for Single Image Dehazing (2018, methodology): “Dark channel prior has been enhanced by optimizing the inherent boundary constraint with weighted L1-norm contextual regularization to estimate transmission map [20].”
Single Image Dehazing via Multi-scale Convolutional Neural Networks (2016, methodology): “Figure 4 shows that the proposed algorithm performs well on each image against the state-of-the-art dehazing methods [1,2,27,28] in terms of PSNR and SSIM.”
Densely Connected Pyramid Dehazing Network (2018, background): “Apart from several works that focus on estimating the atmospheric light [4, 40], most of the other algorithms concentrate more on the accurate estimation of the transmission map and they leverage empirical rule in estimating the atmospheric light [13, 29, 33, 41].”