A lane deviation detection method and system

Abstract

The invention discloses a lane deviation detection method and a system thereof, comprising the following steps: obtaining a distorted image through a forward-looking camera; determining a first lane line and a second lane line in the distorted image, selecting a plurality of first control points on the first lane line and a plurality of second control points on the second lane line; mapping all the first control points into the correction space to obtain a first curve, mapping all the second control points into the correction space to obtain a second curve, and mapping the calibration vanishing points of the front-view camera into the reference vanishing points in the correction space to obtain a reference vanishing point; obtaining the driving area of the vehicle in the correction space according to the reference vanishing point, and judging whether the driving area intersects with the first curve or the second curve, if so, judging the lane deviation to occur, and if not, judging thelane deviation is not to occur. The invention does not need to correct the whole distorted image, greatly saves the calculation amount, reduces the demand for the calculation ability of the equipment, and improves the safety factor of the vehicle driving.

Description

technical field [0001] The invention relates to the field of automobiles, in particular to a lane deviation detection method and system. Background technique [0002] With the increase in the number of driving vehicles and people's emphasis on safe driving, more and more safety monitoring technologies are applied to cars. Lane departure detection, as one of the auxiliary driving technologies, has received more and more attention. The lane deviation detection scheme in the prior art is to use the forward-looking fisheye camera in the panoramic view system to detect the lane deviation. Since the fisheye camera has a large distortion, it is necessary to carry out the distorted image collected by the fisheye camera. For rectification, lane deviation detection is performed after obtaining a general planar image. The prior art needs to rectify the entire distorted image, which requires a large amount of calculation and requires high computing power of the processor. [0003] Ther...

AI Technical Summary

Problems solved by technology

The lane deviation detection scheme in the prior art is to use the forward-looking fisheye camera in the panoramic surround view system to detect the lane deviation. Since the fisheye camera has a large distortion, it is necessary to carry out the distorted image collected by the fisheye camera. Correction, lane deviation detection is performed after obtaining a general planar image. The existing technology needs to correct the entire distorted image, which requires a large amount of calculation and requires high computing power of the processor.

Images