Deep Underwater Image Enhancement

TL;DR

This paper introduces UWCNN, a convolutional neural network that enhances underwater images by directly reconstructing clear images, trained on synthetic data, and outperforming existing methods in real-world scenarios.

Contribution

The paper presents a novel end-to-end CNN model for underwater image enhancement that does not rely on scene-specific parameters and is trained on synthetic datasets for broad applicability.

Findings

Outperforms existing methods both qualitatively and quantitatively

Generalizes well across different underwater scenes

Effective on real-world and synthetic underwater images

Abstract

In an underwater scene, wavelength-dependent light absorption and scattering degrade the visibility of images, causing low contrast and distorted color casts. To address this problem, we propose a convolutional neural network based image enhancement model, i.e., UWCNN, which is trained efficiently using a synthetic underwater image database. Unlike the existing works that require the parameters of underwater imaging model estimation or impose inflexible frameworks applicable only for specific scenes, our model directly reconstructs the clear latent underwater image by leveraging on an automatic end-to-end and data-driven training mechanism. Compliant with underwater imaging models and optical properties of underwater scenes, we first synthesize ten different marine image databases. Then, we separately train multiple UWCNN models for each underwater image formation type. Experimental results on real-world and synthetic underwater images demonstrate that the presented method generalizes well on different underwater scenes and outperforms the existing methods both qualitatively and quantitatively. Besides, we conduct an ablation study to demonstrate the effect of each component in our network.