Underwater image dehaze using scene depth estimation with adaptive color correction

Ding, Xueyan; Wang, Yafei; Zhang, Jun; Fu, Xianping

doi:10.1109/oceanse.2017.8084665

Cited by 55 publications

(15 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Liu et al [11] proposed the deep sparse non-negative matrix factorization (DSNMF) to estimate the image illumination to achieve image color constancy. Ding et al [12] estimated the depth map using the Convolutional Neural Network (CNN) based on the balanced images that were produced by adaptive color correction. Although the above methods can obtain the scene depth map and enhance the underwater image, the deep learning is time consuming.…”

Section: Introductionmentioning

confidence: 99%

A Rapid Scene Depth Estimation Model Based on Underwater Light Attenuation Prior for Underwater Image Restoration

Song

Wang

Huang

et al. 2018

Lecture Notes in Computer Science

202

137

View full text Add to dashboard Cite

Underwater images present blur and color cast, caused by light absorption and scattering in water medium. To restore underwater images through image formation model (IFM), the scene depth map is very important for the estimation of the transmission map and background light intensity. In this paper, we propose a rapid and effective scene depth estimation model based on underwater light attenuation prior (ULAP) for underwater images and train the model coefficients with learning-based supervised linear regression. With the correct depth map, the background light (BL) and transmission maps (TMs) for R-G-B light are easily estimated to recover the true scene radiance under the water. In order to evaluate the superiority of underwater image restoration using our estimated depth map, three assessment metrics demonstrate that our proposed method can enhance perceptual effect with less running time, compared to four state-of-theart image restoration methods.

show abstract

Section: Introductionmentioning

confidence: 99%

A Rapid Scene Depth Estimation Model Based on Underwater Light Attenuation Prior for Underwater Image Restoration

Song

Wang

Huang

et al. 2018

Lecture Notes in Computer Science

202

137

View full text Add to dashboard Cite

show abstract

“…An interesting approach proposed the fusion generative adversarial network to correct colour degradation (Li et al, 2019), while "Water-Net" proposed a generic convolutional neural network to enhance underwater images (Li et al, 2020). The depth estimation can also be important to accurately determine a transmission model for de-hazing an image and can be estimated by a neural network (Ding et al, 2017). (Li et al, 2015) proposed another method for dehazing images including underwater images.…”

Section: Enhancement Of Underwater Imagesmentioning

confidence: 99%

3d Reconstruction and Mesh Optimization of Underwater Spaces for Virtual Reality

Saer¹,

Stentoumis²,

Kalisperakis³

et al. 2020

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. In this contribution, we propose a versatile image-based methodology for 3D reconstructing underwater scenes of high fidelity and integrating them into a virtual reality environment. Typically, underwater images suffer from colour degradation (blueish images) due to the propagation of light through water, which is a more absorbing medium than air, as well as the scattering of light on suspended particles. Other factors, such as artificial lights, also, diminish the quality of images and, thus, the quality of the image-based 3D reconstruction. Moreover, degraded images have a direct impact on the user perception of the virtual environment, due to geometric and visual degenerations. Here, it is argued that these can be mitigated by image pre-processing algorithms and specialized filters. The impact of different filtering techniques on images is evaluated, in order to eliminate colour degradation and mismatches in the image sequences. The methodology in this work consists of five sequential pre-processes; saturation enhancement, haze reduction, and Rayleigh distribution adaptation, to de-haze the images, global histogram matching to minimize differences among images of the dataset, and image sharpening to strengthen the edges of the scene. The 3D reconstruction of the models is based on open-source structure-from-motion software. The models are optimized for virtual reality through mesh simplification, physically based rendering texture maps baking, and level-of-details. The results of the proposed methodology are qualitatively evaluated on image datasets captured in the seabed of Santorini island in Greece, using a ROV platform.

show abstract

“…According to equations (7) and (13), based on the inverse relationship between the scene depth and the transmission map, it can be concluded that the transmission map is approximately zero at the infinity of the scene depth, and the image I T (x, λ) is equal to the background light A (λ). So we can assume that the background light of the image is at the maximum scene depth.…”

Section: Adaptive Background Light Estimatementioning

confidence: 99%

“…Galdran et al [6] proposed a red channel prior model for underwater environments based on dark channel prior and corrected the transmission map in combination with the saturation of the image to achieve natural color correction and visibility improvement. Ding et al [7] estimated the transmission map by scene depth and used the dark channel prior to achieve underwater image dehazing. Finally, the white balance algorithm was used to correct color deviation.…”

Section: Introductionmentioning

confidence: 99%

Underwater Image Enhancement Using Scene Depth-Based Adaptive Background Light Estimation and Dark Channel Prior Algorithms

et al. 2019

View full text Add to dashboard Cite

Due to the complexity of the underwater environment, underwater images captured by optical cameras usually suffer from haze and color distortion. Based on the similarity between the underwater imaging model and the atmosphere model, the dehazing algorithm is widely adopted for underwater image enhancement. As a key factor of the dehazing model, background light directly affects the quality of image enhancement. This paper proposes a novel background light estimation method which can enhance the underwater image. And it can be applied in 30-60m depth with artificial light. The method combines deep learning to obtain red channel information of the background light in the dark channel of the underwater image. Then, the background light is obtained by adaptive color deviation correction. Finally, the experiments of underwater images enhancement are carried out, using the dark channel prior algorithm based on the proposed background light estimation method. The results show that the proposed method effectively improves underwater image blur and color deviation, and is superior to other methods in multiple non-reference image evaluation indicators. INDEX TERMS Adaptive background light estimation, color correction, deep learning, dark channel prior, underwater image enhancement.

show abstract

Underwater image dehaze using scene depth estimation with adaptive color correction

Cited by 55 publications

References 15 publications

A Rapid Scene Depth Estimation Model Based on Underwater Light Attenuation Prior for Underwater Image Restoration

A Rapid Scene Depth Estimation Model Based on Underwater Light Attenuation Prior for Underwater Image Restoration

3d Reconstruction and Mesh Optimization of Underwater Spaces for Virtual Reality

Underwater Image Enhancement Using Scene Depth-Based Adaptive Background Light Estimation and Dark Channel Prior Algorithms

Contact Info

Product

Resources

About