Low-light Image Enhancement via Virtual Exposure and Structural Decomposition

Images taken in low-light environments usually face the problems of insufficient contrast and blurred details. To address these challenges, we propose an efficient and robust low-light image enhancement method, named VESD. Specifically, we first use the camera response model to estimate the exposure and generate an exposure-enhanced image. Subsequently, the original low-light image and the exposure-enhanced image are decomposed into a base layer and a detail layer, respectively, through structural decomposition. Next, we sharpen the detail layer to further highlight the details. Finally, the final enhanced image is generated by fusing the weighted base layer and detail layer. Experimental results show that our method performs well in terms of color vividness, contrast improvement, and detail enhancement. Through qualitative and quantitative experimental evaluations, our method shows stronger competitiveness in all indicators.