In this paper,we design a Linux embedded image acquisition and processing system for mobile robots with mobile trolleys as the experimental platform.The system uses Samsung S3C2440 development board as the core processor and adopts OV9650 camera to collect images.In this paper,the design of the image acquisition module is introduced,and two kinds of label recognition methods-pattern matching and shape feature matching are also introduced.The two methods are compared and analyzed.At the same time,the incline phenomenon of the label appearing in the recognition process is compensated.The embedded image processing system can identify the tag more quickly and accurately,and provide a better experimental platform for the design of mobile robot.%以移动小车为实验平台,设计了一款应用于移动机器人上的Linux嵌入式图像采集处理系统,系统选用三星公司的S3C2440开发板作为核心处理器,采用OV9650摄像头采集图像.重点介绍了图像采集模块的设计,以及两种标签识别方式——模版匹配和形状特征匹配,并对两种方法的进行了对比分析.对识别过程中出现的标签倾斜现象进行了补偿处理.该嵌入式图像处理系统能够较为快捷准确的识别标签,为移动机器人的设计提供了一个较好的实验平台.
展开▼