Research on Block Matching Three-Dimensional Cooperative Filtering Optical Image Denoising Algorithm Based on Noise Estimation
Wenli Zhang, Jingyue Zheng, Kun Liang, Ke-fan Chen, Jianhai Zhao, Jianqiang Liu, Yi Wang, Yuxin Qin · Journal of Nanoelectronics and Optoelectronics · 2021
Optical image denoising is one of the important means to improve the overall clarity of the image or highlight the texture details of the target. However, studies have found that the existing optical image denoising algorithms have a good effect on the images with known noise levels, and have average denoising effects on random noising. Therefore, a novel BM3D optical image denoising algorithm based on noise estimation is proposed. Firstly, the wavelet packet transform is used for the noise estimation of the optical image. Then, according to the noise estimation result, BM3D is used to realize the denoising analysis of the optical image. Finally, PSNR is used to evaluate the effectiveness of the algorithm. Experiments show that the proposed algorithm not only achieves effective denoising of optical image, but also has strong robustness.