Abstract: In order to solve the problems of PCB image edge information missing and carrying a lot of noise in industrial production, the existing de-noising algorithm has poor effect, large amount of calculation and high complexity. Based on this, a PCB image denoising algorithm based on improved NLM is proposed to enhance the denoising quality of PCB image. Firstly, the weight adaptive algorithm based on morphology is used to enhance the PCB image, so that the PCB image retains good edge information. Secondly, the feature matching model is introduced to fuse the enhanced PCB image with the original PCB image. Finally, the PCB image is denoised by improving the weight value of NLM algorithm, and the final denoised image is obtained. Experimental results show that, compared with the existing algorithms, the proposed algorithm retains the edge information of PCB image better, has better denoising effect, significantly improves the image quality, enhances the robustness of the image, improves the calculation speed and reduces the complexity of the algorithm.

Key words: morphological weight adaptive, image enhancement, improved nonlocal mean, feature matching, image denoising