Automatic Red-Channel underwater image restoration

By Adrian Galdran, David Pardo, Artzai Picon, Aitor Alvarez-Gila. Journal of Visual Communication and Image Representation, 2015

932 citations recorded by the source corpus.

New methodology to estimate the depth map in an underwater degraded scene.Novel method to handle the presence of artificial illumination in underwater images.Performance evaluation on images with different illumination and color predominancies. Underwater images typically exhibit color distortion and low contrast as a result of the exponential decay that light suffers as it travels. Moreover, colors associated to different wavelengths have different attenuation rates, being the red wavelength the one that attenuates the fastest. To restore underwater images, we propose a Red Channel method, where colors associated to short wavelengths are recovered, as expected for underwater images, leading to a recovery of the lost contrast. The Red Channel method can be interpreted as a variant of the Dark Channel method used for images degraded by the atmosphere when exposed to haze. Experimental results show that our technique handles gracefully artificially illuminated areas, and achieves a natural color correction and superior or equivalent visibility improvement when compared to other state-of-the-art methods.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/fb393e362434480d47959d146487761be667d4a2

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

1. Chongyi Li
2. Chunle Guo
3. Runmin Cong
4. Wenqi Ren
5. Ming-Hsuan Yang
6. Dacheng Tao
7. Radu Timofte
8. Peixian Zhuang

In Media Technology

1. Chongyi Li
2. Runmin Cong
3. Peixian Zhuang
4. Chunle Guo
5. Wenqi Ren
6. Chunli Zhu
7. Liheng Bian
8. Lintao Peng

Citations in the corpus

• Generalization of the Dark Channel Prior for Single Image Restoration (2018, methodology): “7 shows comparisons of depth estimation based on the DCP [1], DCP variants [12], [13], [24], and our depth”
• A Rapid Scene Depth Estimation Model Based on Underwater Light Attenuation Prior for Underwater Image Restoration (2018, positive): “Since the hazing effect of underwater images caused by light scattering and color change is similar to the fog effect in the air, He’s Dark Channel Prior (DCP) [3] and its variants [4], [8], [9] were used for depth estimation and restoration.”
• Underwater Image Enhancement via Medium Transmission-Guided Multi-Color Space Embedding (2021, background): “Moreover, the accurate estimation of complex underwater imaging parameters challenges current physical model-based methods [3], [4], [28]–[30].”
• An Underwater Image Enhancement Benchmark Dataset and Beyond (2019, methodology): “From left to right are raw underwater images, and the results of Fusion-based [29], Retinex-based [31], Two-step-based [30], UDCP [35], Red Channel [40], Histogram Prior [42], Regression-based [37], Blurriness-based [43], GDCP [38], and reference images.”
• Underwater Single Image Color Restoration Using Haze-Lines and a New Quantitative Dataset (2018, methodology): “Some variations of DCP were proposed for the underwater domain [3], [13], [14], [15], [16].”
• Underwater Image Enhancement With Hyper-Laplacian Reflectance Priors (2022, positive): “Galdran et al. [9] presented an automatic red channel method to exact the dark channel from reversed red and blue-green channels, which recovers colors associated to short wavelengths for compensating the lose of contrast.”
• Underwater Image Enhancement Using a Multiscale Dense Generative Adversarial Network (2020, methodology): “method [8] recovers degraded images by restoring the colors associated with short wavelengths.”
• U-Shape Transformer for Underwater Image Enhancement (2021, methodology): “In addition, physical-model based UIE methods [5, 7, 8, 12, 15, 27, 30, 35, 44] mainly focus on the accurate estimation of medium transmission.”
• Underwater Image Enhancement by Dehazing With Minimum Information Loss and Histogram Distribution Prior (2016, neutral): “Galdran et al. [21] proposed a red channel method that can be considered as a variant of the dark channel prior algorithm, where color associated with short wavelengths is recovered and leads to a restoration of the lost contrast.”
• Using labeled data to evaluate change detectors in a multivariate streaming environment (2009, positive): “ICENet [26] and ICENetv2 [27] show that the two-branch network structure like BiSeNet [11] with a finer spatial branch and a deep context semantic branch is very suitable to segment the river ice with scale diversity characteristic.”
• PUGAN: Physical Model-Guided Underwater Image Enhancement Using GAN With Dual-Discriminators (2023, methodology): “Let’s first review the physical model of the underwater imaging process, which can be expressed as [18]:”
• Real-World Underwater Enhancement: Challenges, Benchmarks, and Solutions Under Natural Light (2019, background): “three major limitations exist in current underwater datasets [3], [6], [10], [13], [20], especially for the purpose of performance evaluation: 1) many existing data sets are unsuitable for evaluating the performance of visibility enhancement, especially those prior model driven methods, as the scene”
• Color Balance and Fusion for Underwater Image Enhancement (2018, background): “[29] observe that, under water, the red component reciprocal increases with the distance to the camera, and introduce the Red Channel prior to recover colors associated with short wavelengths in underwater.”
• Underwater Image Restoration Based on Image Blurriness and Light Absorption (2017, methodology): “) [9], (c) [12], (d) [14], (e) [15], (f) [16], and (g) the proposed method.”