Design Of Road Marking Vehicle Control System Based On Image Recognition

With the rapid development of our economy and the continuous improvement of modernization,China's highway construction has entered a new stage,and road marking has also become an important part of highway construction.As the development of highway construction and maintenance machinery in China is in a lagging state,road marking is mostly manual work,which seriously affects the progress of road marking.Therefore,a new type of intelligent,high-precision,high stability road marking is researched and developed Vehicles are an absolute guarantee for the rapid development of highway construction.This article uses Visual Studio 2012 as the environment platform to perform pre-processing of the image in Open Cv,and selects median filtering to remove noise and sharpen the image;in addition to using,gradient templates in both directions,this The two directions make the image processing range more comprehensive.At the same time,the improved Otsu threshold segmentation algorithm is used to obtain the image threshold value.Finally,the improved Canny algorithm is used to perform road marking images and test the road marking and lane lines in the image to obtain the outer contour edges.In the image,the feature area of the marked line is extracted by the region growth algorithm,and the region of interest is determined by perceptual grouping.The Harris pixel detection algorithm is used to obtain the characteristic pixel points in the region of interest,and the clustering algorithm and the clustering point perception grouping algorithm are used.The characteristic pixel point information is obtained,and a fitting algorithm based on the Hough transform is used to accurately fit the obtained pixel characteristic points,which can accurately determine the deflection direction and deflection angle of the reference line and the road marking vehicle,and feedback the data information The controller controls the steering motor and the driving motor of the road marking vehicle by giving a PWM pulse signal.Complete the design of the software and hardware of the road marking vehicle control system,the production and debugging of the circuit board.Collect the information of road markings and lane lines in road images,temperature sensors,encoders and other return values and perform analog-to-digital conversion and data calculation in the main control chip,and at the same time,the motor rotation performance and the accuracy of the motor control system rotation angle Test,transmit data to the host computer through Wi-Fi communication,display the collected data and detection results on the human-computer interaction interface built by the host computer,and realize the reading,analysis and storage of relevant parameters based on the data.After testing,it can be seen that the image processing of road images has a better effect on the extraction of road markings,lane lines and road reference lines.The control system of road marking vehicles can achieve stable control while managing the data.Lay the foundation for further automation.