Research On Lane Recognition And Departure Warning In Monocular Vision

With the development of road traffic and the increasing number of cars,the issue of traffic safety has attracted attention.The incidence of traffic accidents is also increasing year by year.Statistics show that most of the traffic accidents are caused by the driver's tired driving and relaxation of vigilance.In response to the current problems,combined with the rapid development of today's technology,automobile assisted driving technology has emerged at the historic moment.Its role is mainly to assist the driver to give early warning to the driver unconsciously to prevent traffic accidents.This article makes in-depth research and analysis on the key lane detection and departure warning technology in assisted driving.The most important of the lane departure warning system studied in this paper should be the recognition and tracking of lane lines.The system first uses the on-board camera to obtain the road driving information of the vehicle,provides data for the system to make a judgment,and whether the vehicle will deviate from the current driving lane.,Whether to give the driver a warning signal in the end.This article firstly performs preprocessing operations on the extracted lane line information,mainly including grayscale processing of interference information.In the denoising and filtering,a practical median filter is selected,and the histogram equalization is used in image enhancement.Road information,compared with two commonly used methods in binarization,improved on the basis of the original maximum between-class variance method.The second is the edge detection of the road,which is improved on the basis of the Canny edge detection algorithm to make its detection results more accurate.In the process of lane recognition,this paper selects the lane line detection algorithm based on the improved Hough transform of the highway model,which can better fit the lane line.Finally,the car camera is calibrated in order to convert from the image coordinate system to the world coordinate system.After comparing and analyzing several current lane departure warning models,the TLC model which is relatively easy to implement is chosen as the lane departure warning model.The experimental results show that the lane recognition and departure warning methods proposed in this paper meet the actual requirements and achieve the ideal results.