Xiaofang Li

Low-light image enhancement aims to restore visually pleasing normal-light images from degraded low-light observations. Most existing methods handle luminance variation from the enhancement perspective. As a result, the degradation process from a normal-light image to a low-light observation is usually not explicitly characterized. In addition, degradation-oriented optimization is often computationally expensive due to repeated iterative updates. To address these issues, based on the alternating direction method of multipliers (ADMM), a degradation-oriented step-size-consistent unfolding network SCAUNet is proposed. Specifically, a low-light image is modeled as the element-wise product of a normal-light image and a luminance degradation operator, together with additive noise. Based on this formulation, low-light enhancement is converted into the joint estimation of the target image and the degradation operator. Then, a state-based one-step ADMM solver is developed, and a step-size consistency constraint is introduced to improve the reliability of one-step unfolding. Extensive experiments on LOL-v1 and LOL-v2 demonstrate the effectiveness of the proposed SCAUNet. Compared with existing state-of-the-art methods, SCAUNet yields better enhancement quality, especially in preserving image structures, correcting illumination, and suppressing artifacts. Strong generalization ability is also verified on four no-reference low-light datasets, and promising results are obtained on single-image exposure correction.

In this paper, a novel active contour model for image segmentation is proposed, which comprises two components: edge energy and region energy. First, a preprocessing step is performed to obtain a denoised smooth image using an off-the-shelf denoiser, which effectively removes noise and unwanted edges. Then we calculate the gradient of the preprocessed image for the edge detection function. In addition, our model considers both local and global information of the image in the region energy. Specifically, we construct a constraint to minimize the differences between the denoised smooth image and the local fitted image, and design fitting terms of the image intensity inside and outside the target boundary. Extensive experiments indicate that our proposed model can effectively segment images even in the presence of noise and intensity inhomogeneity. Moreover, the segmentation accuracy and robustness of this model are better than those of relevant segmentation models.