Vehicle reverse positioning method based on monocular vision and utilizing road lane lines

Abstract

The invention discloses a monocular vision-based vehicle reverse positioning method using a road lane line, comprising the steps of collecting a road image and performing distortion correction, lane line detection and straight line approximation, constructing a road plane positioning model based on a road plane hypothesis model, positioning lane lines and processing and positioning to obtain the distance between a vehicle and left and right lane lines and the included angle between the vehicle and the lane lines, and completing vehicle positioning by using the lane lines. The invention provides a road plane positioning model and a corresponding positioning frame aiming at the difficult problem of monocular lane line positioning, realizes the positioning of the lane line and the relative position of a vehicle by a monocular single-frame image, and solves the problem of uncertain monocular vision scale through pre-calibrated camera parameters, so that accurate scale information is obtained, and the vehicle positioning precision is improved.

Description

technical field [0001] The invention belongs to the technical field of vehicle positioning, and more specifically relates to a monocular vision-based vehicle reverse positioning method using road lane lines, which is used to determine the position of the vehicle itself in the lane. Background technique [0002] In unmanned driving, obtaining the vehicle's own positioning is a prerequisite for perceptual decision-making, and the positioning system needs to have better robustness and higher accuracy. Sensors commonly used for positioning include cameras, lidar, global GPS positioning, etc. Among them, monocular cameras have important use value in unmanned driving. The monocular camera has high-quality information, and does not require external reference calibration, and saves a lot of computing resources compared with multi-eye vision and lidar; at the same time, the monocular camera does not rely on external signals, and has better performance than GPS. robustness. Unfortun...

AI Technical Summary

Problems solved by technology

Although this technical solution can meet the needs of unmanned driving in terms of speed, it is highly dependent on traffic signs. At the same time, the coordinates of the control points have certain requirements for the selection of traffic signs, and the positioning function is lost in places without road signs.

Images