Inter-color correlation-based stereo matching sparse block aggregation strategy method

Abstract

The invention relates to an inter-color correlation-based stereo matching sparse block aggregation strategy method. The method is characterized by comprising the following steps of: obtaining a groupof test images from a Middlebury data set platform and transmitting the test images into a computer 1; improving a matching cost function by utilizing correlation between a pixel and a color space soas to better express local properties of the pixel; constructing a mean value block-based sparse aggregation strategy to improve the algorithm efficiency; and eliminating mis-matched points by utilizing left-right consistency detection and median filter. Through the method, the matching precision in weak texture areas is improved and the operation time is shortened.

Description

Technical field [0001] The invention relates to a three-dimensional matching sparse block aggregation strategy method based on color correlation, and belongs to the technical field of computer vision. Background technique [0002] Stereo matching is one of the most active research topics in the field of computer vision, and it has a wide range of applications in three-dimensional reconstruction, unmanned driving, virtual reality and other fields. Its purpose is to find the corresponding pixel point of each pixel in another image from a given pair of left and right images. The difference between the positions of two matching pixels is called parallax. [0003] The current mainstream local stereo matching algorithms are divided into adaptive window algorithm and adaptive weight algorithm. The adaptive window algorithm uses different window sizes for different pixels according to the local attributes of the pixels. In contrast, the adaptive weighting algorithm adjusts the weights of...

AI Technical Summary

Problems solved by technology

Hu W defines a virtual support window for a pixel in the matching cost aggregation, which can select the most relevant pixel, effectively avoiding the mismatch caused by the uncertainty of image correspondence, but the amount of calculation is still large

[0004] In the existing stereo matching process, when the size of the matching window is smaller, the stereo matching accuracy in the strong texture area of ​​the image is higher, but due to the lack of sufficient information, the parallax value error in the single texture area will increase

Conversely, when the size of the matching window is larger, the interference of image noise can be eliminated more effectively, but because it contains more information, it will lead to blurred boundaries at discontinuous areas of parallax, and the calculation of the algorithm is larger

Images