Parallax image generation method and system based on binocular stereo vision matching

Abstract

The invention provides a parallax image generation method and system based on binocular stereo vision matching. The parallax image generation method comprises the steps of obtaining a pair of speckleimages, wherein the pair of speckle images are images of speckles captured by an object through a pair of binocularly calibrated cameras under the irradiation of a set light beam; inputting the pair of speckle images into a preset deep learning neural network to generate corresponding parallax images, wherein the deep learning neural network is obtained by training a plurality of pairs of speckleimages. According to the method and the system, a pair of speckle images is used as input, a parallax image is directly output through binocular vision stereo matching network (deep learning neural network) processing, end-to-end calculation of the parallax image and an end-to-end network design framework are realized, post-processing steps of stereo matching, such as sub-pixel enhancement, filtering and noise reduction, refinement and the like, are eliminated, and the operation overhead is greatly reduced.

Description

technical field [0001] The invention belongs to the field of image processing, and more specifically relates to a parallax image generation method, system, electronic equipment and readable medium based on binocular stereo vision matching. Background technique [0002] The rapid development of depth sensors has brought new changes to the field of computer vision, and depth estimation information has been fully applied to solutions to problems related to autonomous driving, 3D reconstruction, behavior recognition, and liveness detection. At present, mainstream depth sensors mainly include four types of technologies: time of flight (ToF), lidar, structured light and binocular stereo vision. [0003] The technical steps of binocular stereo vision include: camera calibration to obtain camera parameters; epipolar line correction to make the epipolar lines of the left and right cameras horizontal; image pixel matching after correction to obtain disparity map; depth information obt...

AI Technical Summary

Problems solved by technology

Therefore, the current binocular stereo matching method has the following shortcomings: Although the CNN-based matching method improves the matching speed and accuracy, it does not solve the essential problems of binocular stereo matching. area, it is impossible to accurately find matching points, and post-processing steps for stereo matching are required, such as sub-pixel enhancement, filter noise reduction, refinement, etc., the process is cumbersome

Images