Underwater Image Restoration Based on Convolutional Neural Network

Yan Hu, Keyan Wang, Xi Zhao, Hui Wang, Yunsong Li
Proceedings of The 10th Asian Conference on Machine Learning, PMLR 95:296-311, 2018.

Abstract

Restoring degraded underwater images is a challenging ill-posed problem. Existing priors-based approaches have limited performance in many situations due to their hand-designed features. In this paper, we propose an effective convolutional neural network (CNN) based approach for underwater image restoration, which consists of a transmission estimation network (T-network) and a global ambient light estimation network (A-network). By learning the relationship between the underwater scenes and their corresponding blue channel transmission map and global ambient light respectively, we can recover and enhance the underwater images with an underwater optical imaging model. In T-network, we use cross-layer connection and multi-scale estimation to prevent halo artifacts and to preserve edge features. Moreover, we develop a new underwater image synthetic method for training, which can simulate underwater images captured in various underwater environments. Experimental results of synthetic and real images demonstrate that our restored underwater images exhibits more natural color correction and better visibility improvement against these state-of-the-art methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v95-hu18a, title = {Underwater Image Restoration Based on Convolutional Neural Network}, author = {Hu, Yan and Wang, Keyan and Zhao, Xi and Wang, Hui and Li, Yunsong}, booktitle = {Proceedings of The 10th Asian Conference on Machine Learning}, pages = {296--311}, year = {2018}, editor = {Zhu, Jun and Takeuchi, Ichiro}, volume = {95}, series = {Proceedings of Machine Learning Research}, month = {14--16 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v95/hu18a/hu18a.pdf}, url = {https://proceedings.mlr.press/v95/hu18a.html}, abstract = {Restoring degraded underwater images is a challenging ill-posed problem. Existing priors-based approaches have limited performance in many situations due to their hand-designed features. In this paper, we propose an effective convolutional neural network (CNN) based approach for underwater image restoration, which consists of a transmission estimation network (T-network) and a global ambient light estimation network (A-network). By learning the relationship between the underwater scenes and their corresponding blue channel transmission map and global ambient light respectively, we can recover and enhance the underwater images with an underwater optical imaging model. In T-network, we use cross-layer connection and multi-scale estimation to prevent halo artifacts and to preserve edge features. Moreover, we develop a new underwater image synthetic method for training, which can simulate underwater images captured in various underwater environments. Experimental results of synthetic and real images demonstrate that our restored underwater images exhibits more natural color correction and better visibility improvement against these state-of-the-art methods.} }
Endnote
%0 Conference Paper %T Underwater Image Restoration Based on Convolutional Neural Network %A Yan Hu %A Keyan Wang %A Xi Zhao %A Hui Wang %A Yunsong Li %B Proceedings of The 10th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2018 %E Jun Zhu %E Ichiro Takeuchi %F pmlr-v95-hu18a %I PMLR %P 296--311 %U https://proceedings.mlr.press/v95/hu18a.html %V 95 %X Restoring degraded underwater images is a challenging ill-posed problem. Existing priors-based approaches have limited performance in many situations due to their hand-designed features. In this paper, we propose an effective convolutional neural network (CNN) based approach for underwater image restoration, which consists of a transmission estimation network (T-network) and a global ambient light estimation network (A-network). By learning the relationship between the underwater scenes and their corresponding blue channel transmission map and global ambient light respectively, we can recover and enhance the underwater images with an underwater optical imaging model. In T-network, we use cross-layer connection and multi-scale estimation to prevent halo artifacts and to preserve edge features. Moreover, we develop a new underwater image synthetic method for training, which can simulate underwater images captured in various underwater environments. Experimental results of synthetic and real images demonstrate that our restored underwater images exhibits more natural color correction and better visibility improvement against these state-of-the-art methods.
APA
Hu, Y., Wang, K., Zhao, X., Wang, H. & Li, Y.. (2018). Underwater Image Restoration Based on Convolutional Neural Network. Proceedings of The 10th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 95:296-311 Available from https://proceedings.mlr.press/v95/hu18a.html.

Related Material