A Lane Departure Early Warning Method Independent of Camera Parameters and Vehicle System

Abstract

The invention discloses a lane departure early warning method independent of camera parameters and a vehicle system, which aims at a problem that a lane departure early warning method based on image information combined with a lane model can not realize early warning or has strong dependence on the camera parameters and the vehicle system. The lane departure early warning method comprises the following steps of firstly calculating the transverse distance between the head of a vehicle and the middle line of the same reference lane by utilizing a CCP (car current position) lane departure decision method, and calling the transverse distance as ccpData; and secondly predicting the ccpData within a future period by utilizing a grey prediction method according to a group of ccpData within the past period, judging whether lane departure generates or not at the future moment, and early sending out departure warning. According to the lane departure early warning method independent of the camera parameters and the vehicle system, disclosed by the invention, the CCP method establishes the foundation for the independence of the camera parameters and the vehicle system of the early warning method, the momentat which a lane departure eventprobably generates in the future is estimated by utilizing the grey prediction method, and the lane departureearly warning function is realized.

Description

technical field [0001] The present invention mainly relates to the field of lane departure warning based on vision (Lane Departure Warning based on Vision), in particular to a lane departure warning method based on CCP (Car's Current Position) and gray prediction. Background technique [0002] With the rapid development of the economy, people's demand for cars continues to rise, and the incidence of traffic accidents also increases rapidly. Safe driving has also become one of the focuses of social attention. Among all traffic accidents, the proportion of lane departure accidents caused by driver inattention is as high as 37.4%. Therefore, how to reduce the incidence of traffic accidents caused by vehicle deviation has become a major challenge in today's society. Surveys show that if early warning can be given 0.5 seconds in advance, the incidence of traffic accidents will be reduced by 60%. Lane departure warning technology is an effective solution to this problem. Lane d...

AI Technical Summary

Problems solved by technology

However, this method generally establishes the geometric model of the image plane and the real space, requires camera parameters and vehicle driving status, and is highly dependent on the camera and vehicle system, which is not easy to achieve popularization; in this type of method, the camera and vehicle system The weakly dependent CCP method can only judge whether the vehicle has left the lane at the current moment, but it cannot provide early warning of the occurrence of the lane departure event, so it cannot provide enough warning time for the driver to take necessary measures to correct the lane departure phenomenon

Images