A Switched View of Retinex: Deep Self-Regularized Low-Light Image Enhancement

Self-regularized low-light image enhancement does not require any normal-light in training, thereby freeing from the chains of paired or unpaired training data that are time-consuming to obtain. However, existing methods suffer color deviation and fail generalize various lighting conditions. This paper presents a novel self-regularized method based on Retinex, which, inspired by HSV, preserves all colors (Hue, Saturation) only integrates Retinex theory into brightness (Value). Besides, we design random disturbance approach generate another abnormal same scene. It is combined with original form estimate reflectance, which achieved CNN. The assumed irrelevant illumination according theory, treated as enhanced brightness. Our efficient decoupled two subspaces, i.e., brightness, for better preservation enhancement. Extensive experiments demonstrate our outperforms multiple state-of-the-art algorithms qualitatively quantitatively adapts more code available at https://github.com/Github-LHT/A-Switched-View-of-Retinex-Deep-Self-Regularized-Low-Light-Image-Enhancement.