(毕业设)飞思卡尔智能车及机器视觉.docx

图像处理在智能车路径识别中的应用摘要机器视觉技术在智能车中得到了广泛的应用，这项技术在智能车的路径识别、障碍物判断中起着重要作用。基于此，依据飞思卡尔小车的硬件架构,研究机器视觉技术应用于飞思卡尔小车。飞思卡尔智能车处理器采用了MC9S12XS128芯片，路况采集使用的是数字摄像头OV7620。由于飞思卡尔智能车是是一款竞速小车，因此图像采集和处理要协调准确性和快速性，需要找到其中的最优控制。因此本设计主要需要完成的任务是：怎样用摄像头准确的采集每一场的图像，然后怎样进行二值化处理；以及怎样对图像进行去噪处理；最后也就是本设计的难点也是设计的核心，怎样对小车的轨迹进行补线。本设计的先进性，在众多的图像处理技术中找到了适合飞思卡尔智能车的图像处理方法。充分发挥了摄像头的有点。经过小车的实际测试以及相关的MATLAB仿真，最终相关设计内容都基本满足要求。小车的稳定性和快速性得到显著提高。关键词：OV7620，视频采集，图像处理，二值化The Application of Image Processing in the Recognition of Intelligent Vehicle PathABSTRACTCameraMachine vision technology in the smart car in a wide range of applications, the technology identified in the path of the smart car, and plays an important role in the obstacles to judge. Based on this, based on the architecture of the Freescale car, machine vision technology used in the Freescale car. Freescale smart car the processor MC9S12XS128 chip traffic collected using a digital camera OV7620.Freescales Smart car is a racing car, so the image acquisition and processing to coordinate the accuracy and fast, you need to find the optimal control. This design need to complete the task: how to use the camera to accurately capture every image, and then how to binarization processing; and how to image denoising; last is the difficulty of this design is the design of the core, how to fill line on the trajectory of the car.The advanced nature of the design found in many image processing techniques of image processing methods for Freescale Smart Car. Give full play to the camera a bit. The actual testing of the car and MATLAB simulation, the final design content can basically meet the requirements. The cars stability and fast to get improved significantly.KEY WORDS:OV7620，Video Capture，PictureProcessing，Binarization目　录前　言1第1章飞思卡尔赛车及机器视觉的概述21.1 智能车的研究背景21.1.1 智能车的发展历史21.1.2 应用前景21.2 智能车设计要求介绍31.3 机器视觉介绍41.4 小结4第2章主要思路及技术方案概要52.1 总体设计主要方法步骤52.2 摄像头的对比与选择52.2.1 摄像头的选取52.2.2 模拟摄像头62.2.3 数字摄像头62.2.4 摄像头的选定72.3 二值化方案的选取72.3.1 双峰值法72.3.2 迭代法82.3.3 大津法82.3.4 灰度拉伸-一种改进的大津法