The Research And Exploitation Of Vehicle Door Open Accident Warning System

According to relevant data,the proportion of accidents caused by lane departures has increased year by year.This shows that it is very important to study the lane departure warning system.The lane departure warning system can help the driver to concentrate and prevent the car from deviating from the lane,thereby reducing traffic accidents and reducing the casualty rate.This paper takes the automobile as the research object,and combines the related theoretical knowledge of computer vision,image preprocessing and deviation warning algorithm to design a lane departure warning system based on image processing.Research work includes:(1)Estimation of parameters inside and outside the camera.Firstly,by comparing the traditional calibration method with the self-calibration method,this paper finally selects the Zhang Zhengyou calibration method in the traditional calibration method.Then made an 8×7 black and white interlaced checkerboard calibration board and selected a camera.On this basis,through the multi-angle shooting of the calibration plate,and using the Matlab software to obtain the internal parameters of the camera.Finally,by adjusting the installation height and angle of the camera multiple times,the external parameters of the camera are obtained.(2)Image preprocessing.Firstly,a customized camera is used to capture the road in real time,and then the captured lane line is taken as the main target of image preprocessing,and the region of interest is set and grayscaled.This laid the foundation for subsequent system target detection.(3)Extraction of feature points.After image preprocessing,there is still a lot of redundant information in the image.In order to make the research objectives clearer,feature points must be extracted.In view of this,this paper uses two methods of threshold processing and edge detection,and then through Matlab simulation experiment,it is found that it has achieved good expected results.(4)Lane line detection.Based on image processing,this paper performs target recognition detection on straight and curved roads.For the straight detection,this paper first establishes a linear model,and then uses the Matlab software to identify the lane line based on the Hough variation.For the curve detection,this paper firstly designs a curve recognition model based on the center point extraction,and then simulates the collected road video through Matlab software.The experimental results show that the simulation results are better,can meet the requirements of the system and have better stability.(5)Simulation and analysis of the results of the early warning system.Firstly,several classic vehicle departure warning models are briefly introduced.Then,based on the comprehensive analysis,an early warning model based on lateral distance is established.Based on the distance between the center line of the car and the center line of the lane,the current state of the vehicle is used for early warning decisions.At the same time,other programs required in the simulation experiment were designed,including ranging speed measurement program and execution alarm program.Finally,Matlab software was used to simulate the road video captured in real time.The simulation results show that when the car is in a state of being deviated or has deviated,the system can prompt the driver to reach the expected warning effect and the false alarm rate is low.At the same time,the system also meets the requirements of real-time,which fully proves the feasibility and accuracy of the system.