Stereo matching optimization method for binocular vision system

Abstract

The invention provides a stereo matching optimization method of a binocular visual system. The method comprises a step of using the binocular visual system to gain a binocular image; a step of correcting a distortion and polar line of the binocular visual system; a step of transferring the binocular image from a colorful image into a grayscale image; a step of regarding one of the grayscale images as a basic standard and matching with another image according to pixel masterplate; a step of optimizing ZNCC factor; a step of regarding the optimized ZNCC as a similarity measure to calculate in a matched mode so as to ensure matching parallax of each pixel point; a step of producing a dense disparity map according to the matching parallax of each pixel point. The method has the advantages of reducing times of calculation greatly and reducing calculating load of computer chip greatly by optimizing a computational formula of the ZNCC factors and calculating method. The method can improve real-time performance of the stereo matching as well as reducing cost of day-to-day and military high-precision image identification.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to a stereo matching optimization method of a binocular vision system. Background technique [0002] With the development of computer vision, stereo vision technology is widely used in robot navigation, intelligent transportation, military guidance and so on. Stereo vision is to use the disparity value of the same point in space on the two camera planes to calculate the three-dimensional coordinates of the space point, and the disparity must be obtained through stereo matching. Therefore, stereo matching is the most important and difficult step in stereo vision measurement. one. According to different matching primitives, stereo matching methods can be roughly divided into two types: region-based and feature-based stereo matching. The feature-based stereo matching method is to selectively extract features and perform matching. Due to the sparsity and irregularity of features, a dens...

AI Technical Summary

Problems solved by technology

However, due to the relatively complex calculation factors of ZNCC, it requires high calculation costs to perform correlation calculations on all pixels in the image, and it is necessary to use a calculation chip that can carry higher calculation loads, which greatly increases the application cost, and the speed of stereo matching is also slow. As ideal as possible, it cannot meet the requirements of fast driving and military guidance applications that require high-speed identification

[0004] In the prior art, in order to improve the matching speed and ensure that the calculation characteristics of the matching template in the entire image remain unchanged, the improvement method for the ZNCC calculation factor itself is to expand its calculation formula and consider the template in the adjacent pixel matching process. Translational characteristics, using the BOX filter method to optimize the correlation calculation of the expanded part and reduce redundant calculations in the matching process, but these methods fail to fully simplify and integrate the ZNCC calculation factors, and at the same time fail to fully consider the calculation process The overall and internal correlation of the template in the middle is still not separated from the framework of point-by-point calculation of each point on the pixel calculation template, and the complexity of the calculation increases with the size of the template. For higher-precision stereo matching, the computing chip The greater the requirements and the stereo matching speed, the greater the limitation, which is not conducive to its application in daily and military fields when high-precision image recognition is required

Images