A bilateral filtering acceleration method based on factorization and mean square error optimization

Abstract

The invention provides a bilateral filtering acceleration method based on factorization and mean square error optimization, and the method comprises the following steps: inputting an image to be filtered, and calculating an accumulated histogram of the image; performing factorization on the value domain filtering kernel function to obtain a filtering coefficient and an approximation function, performing mean square error optimization, and sampling on the two-dimensional grid to obtain a matrix; performing singular value decomposition on the matrix to obtain a characteristic value and a characteristic vector; performing interpolation operation on the feature vector to obtain a complete function; and according to the obtained filtering coefficient and the approximation function, performing iterative summation to realize rapid bilateral filtering, and outputting a filtering result graph.

Description

technical field [0001] The invention relates to an image processing technology, in particular to a bilateral filtering acceleration method based on factorization and mean square error optimization. Background technique [0002] With the development and popularization of mobile networks and smart phones, the acquisition and transmission of image resources is easier and more accessible. As the carrier of information, images appear more and more in our daily life, and the pursuit of high-quality image information is becoming more and more important. It is becoming more and more common, so the bilateral filter with edge-preserving denoising ability (while denoising can keep the edge details from being blurred) has been more and more widely used: in computer graphics, computer vision and computer photography and other disciplines Popular filter; versatile tool in areas such as denoising, non-destructive image enhancement, video abstraction, and stereo matching; can even be used a...

AI Technical Summary

Problems solved by technology

[0003] However, since the bilateral filter itself is a nonlinear weighted mean filter, it can only be realized by simple point-by-point calculation, so it is very time-consuming for high-latitude and high-resolution images, so the bilateral filter cannot be used in real-time scenes

In order to solve this problem, more and more scholars are committed to speeding up the bilateral filter. Although some achievements have been made, it still cannot meet the needs of practical applications. The main challenge is how to find the best compromise between speed and accuracy.

Images