Design Of An Online Detection System For Vehicle Deviation

Driving deviation is a common vehicle problem.When it happens,not only cause serious abrasion of vehicle parts,but also increase the uncontrollability to the driver.At present,there is no such system that can fully and automatically detect the driving deviation of the vehicle scientifically and accurately.In order to prevent the problematical products from entering the market,a system of automatic online detection of vehicle deviation using binocular vision is designed.Research the algorithms of binocular vision principle to measure vehicle deviation,including camera calibration,image preprocessing,feature detection,stereo matching.First of all,expound the research background and significance of the measurement of vehicle deviation.Describe the phenomena and hazards of the vehicle deviation.Introduce the research status of the measurement technique of the vehicle deviation inland and abroad.Propose the main research content.Design the on-line imaging detection system for vehicle deviation.According to the detection requirements,discuss the hardware scheme,the selection basis and the structure of the measurement software.Then,research the principle of stereo vision ranging system.After lens distortion correction,calibrate the binocular system with a dot matrix calibration plate to acquire the internal and external parameters.In order to make the vehicle's key features immune to noise,research the noise reduction algorithm.After the image grayscale processing,enhance and sharpe the image to obtain a more clear and distinct vehicle feature.Then,research the feature extraction and stereo matching algorithm.Use the edge detection operator to extract the vehicle contour.Use the Huff transform to detect the lane.The SURF algorithm can extract the feature points of the car body for image stereo matching.After eliminating the abnormal matching pairs,obtain the parallax image.And test the minimum depth resolution of the system.Subsequently,according to the requirements of GB12534-90,build hardware and software systems and test the system outdoors on the road.Finally,the binocular system measured the depth,deviation,deviation rate,deviation angle and other information.Compare them with the manual measurement results to verify the system's measurement capabilities.The error of deviation measurement is less than 0.1m,which is consistent with the result of manual measurement.These experiments verify the effectiveness of the system.