Brightening and denoising lowlight images
Xinwei Yang, Xiangbo Lin · 2015
This paper aims to improve the quality of images in low light conditions. After nonlinear intensity stretching, the enlarged mixed noise is reduced using a newly proposed low rank denoising algorithm. By similar patch stacking, the established noisy matrices are assumed to be composed of low-rank noise free image, sparse random impulse noise and zero-mean Gaussian noise. Minimizing the rank of the matrices, the mixed noise can be reduced efficiently. Comparative experiments on synthetic and real data are used to verify the algorithm's performance.