Abstract：Underwater images captured in diverse underwater scenes exhibit varying types and degrees of degradation, including color deviations, low contrast, blurry details, etc. Single image enhancement methods tend to insufficiently address the diverse degradation issues, resulting in inappropriate results that do not align well with human visual perception or underwater color prior. To overcome these deficiencies, we develop a novel reinforcement learning framework that selects a sequence of image enhancement methods and configures their parameters in a self-organized manner for the purpose of underwater image enhancement. In contrast to end-to-end deep learning-based black-box mechanisms, the novel framework operates in a white-box fashion where the mechanisms for the method selection and parameter configuration are transparent. Furthermore, our framework incorporates the human visual perception and the underwater color prior into non-reference score increments for rewarding the underwater image enhancement. This breaks through the training limit imposed by volunteer-selected enhanced images as references. Comprehensive qualitative and quantitative experiments ultimately demonstrate that our framework outperforms nine state-of-the-art underwater image enhancement methods in terms of visual quality, and achieves better performance in five underwater image quality assessment metrics on three underwater image datasets. We release our code at https://gitee.com/wanghaoupc/Self_organized_UIE .
KeyWord：Self-organized; Underwater image enhancement; Human visual perception; Underwater color prior; Reinforcement learning;

相关文献：
1.Improving PRISMA hyperspectral spatial resolution and geolocation by using Sentinel-2: development and test of an operational procedure in urban and rural areas
2.Gap completion in point cloud scene occluded by vehicles using SGC-Net
3.A flexible trajectory estimation methodology for kinematic laser scanning
4.Navigating the publishing landscape in times of revolutionary changes
5.Global Streetscapes — A comprehensive dataset of 10 million street-level images across 688 cities for urban science and analytics
6.A semi-supervised multi-temporal landslide and flash flood event detection methodology for unexplored regions using massive satellite image time series
7.EMET: An emergence-based thermal phenological framework for near real-time crop type mapping
8.Remote sensing vegetation Indices-Driven models for sugarcane evapotranspiration estimation in the semiarid Ethiopian Rift Valley
9.CodeUNet: Autonomous underwater vehicle real visual enhancement via underwater codebook priors
10.Towards a gapless 1 km fractional snow cover via a data fusion framework