基于单片机的智能车图像处理与道路识别算法研究

摘要

以MC9S12XS128单片机为硬件支撑,以MATLAB 7.10.0为软件平台,构建了智能车工作的整体框架,并着重研究了智能车的图像采集、图像处理及道路识别控制算法.智能车控制系统中,采用OV7620摄像头采集道路信息,通过二值法、中值滤波法对采集的道路信息进行识别处理,进而提取道路中心线,以此作为智能车的方向引导线,根据此引导线的斜率进行道路类型的判断识别.试验结果表明,智能车能够准确地识别道路类型,并且在保证稳定性的前提下高速行驶.%The whole framework of the intelligent vehicle is constructed with MC7S12XS128 singlechip as the hardware and MATLAB 7.10.0 as the software platform.The image acquisition,image processing and road recognition control algorithms of the intelligent vehicle were studied emphatically.In the intelligent vehicle control system,the road information was collected by OV7620 camera,and the road information was recognized and processed by binary method and median filtering method and then the road centerline was extracted.The centerline was taken as the guide line to direct the smart car while the offset of the guide line was taken for identification of the road types.The test results show that the smart car can accurately identify the type of road and can run steadily at high speed.