Visual positioning method based on robust feature tracking

Abstract

The method discloses a robust feature tracking and stereoscopic vision positioning technology based on image processing and machine vision. The technology can integrate inertial information and visual information and achieve reliable stereoscopic vision positioning under camera waggling conditions and outdoor light conditions. Images are collected through a binocular video camera in real time, and rotation information of the camera is collected with an inertial measurement unit. Feature points in the images are extracted with a feature extraction algorithm, and the feature points of the left image and the feature points of the right images are matched stereoscopically. The inertial information is combined and the inertia and the KLT algorithm are integrated to track the feature points, so that the reliability of the feature tracking is promoted. Three-dimensional information of the feature points is restored according to the double vision geometric principle. Motion parameters of the camera are obtained through position information of the feature points with the Gaussian and Newton iteration method. The accuracy of visual positioning is further promoted with the RANSIC algorithm. The whole process is iterated continuously, and thus real-time calculation of the posture and the position of the camera is achieved.

Description

Technical field: [0001] The invention is based on the robust feature tracking and stereo vision positioning technology in the related fields of image processing and machine vision. This technology can use inertial information and visual information as input to achieve reliable stereo vision positioning under camera shaking and outdoor lighting conditions. Background technique: [0002] In the field of computer vision, binocular stereo vision is an important research direction and has wide applicability. Binocular stereo vision can be used for self-positioning when the mobile robot performs tasks autonomously and in the assisted travel positioning system for the blind. In these systems, the accuracy and speed of positioning directly affect the performance and efficiency of the entire system. However, the existing visual positioning methods have certain defects: [0003] 1) When the camera shakes or rotates violently during the image acquisition process and cannot work in a...

AI Technical Summary

Problems solved by technology

[0008] The present invention is mainly used to solve the problem of unreliable positioning of existing visual positioning technology in camera shaking and outdoor lighting environment, and proposes a visual positioning algorithm that combines inertial information and visual information, and predicts the current moment of feature points by adding inertial information Iterative initial position in the image, combined with an affine photometric motion model for robust feature tracking, thereby improving the reliability of visual localization

Images