Video Noise Reduction Method and System

Abstract

The present invention provides a method and system for video noise reduction. The method includes the steps of: acquiring multiple frames of video images, and dividing the acquired video images into blocks; generating an initial matrix according to each image block obtained by dividing into blocks; determining an initial matrix according to the initial matrix The mean value block corresponding to the matrix; the PCA projection matrix is ​​obtained by performing PCA transformation on the initial matrix; the residual matrix is ​​obtained according to the mean value block and each of the image blocks; the residual matrix of each image block is obtained by dimensionality reduction through the PCA projection matrix respectively The feature block of each image block; use the feature block to perform block matching to obtain the weight value of the reference feature block relative to the current feature block; use the weight value to perform weighted filtering on the current image block, and use the filtered image block to form a filtered video image. By adopting the scheme of the invention, both the noise reduction efficiency and the noise reduction effect can be improved.

Description

technical field [0001] The present invention relates to the field of video technology, in particular to a video noise reduction method and system. Background technique [0002] In the field of video technology, especially in the field of video surveillance, in the case of low ambient brightness, the surveillance video will be accompanied by more noise, and the visual effect is poor. Especially when the original video is enhanced through brightness enhancement, the video noise will also be enhanced, which is especially serious and seriously affects the visual effect. The traditional image noise reduction method is only for a single image, and the noise reduction effect is poor when applied to video files, and the calculation complexity is high. In recent years, some time-space domain joint noise reduction methods for video images have emerged, which use motion estimation in the time domain to search for the best matching image blocks, and use the similarity between matching ...

AI Technical Summary

Problems solved by technology

This method needs to use image block matching in both the time domain and the air domain, which has high computational complexity and low noise reduction efficiency, and cannot meet the real-time requirements in the field of real-time video surveillance. Moreover, due to the existence of noise in image blocks, often Causes mismatching, and the mismatching rate is higher when the noise is serious, resulting in poor noise reduction effect

Images