一种基于ARM的割草机器人边界图像处理方法设计研究

摘要

机器视觉是割草机器人边界图像识别的关键技术之一,采用ARM系列STM32F103ZE和OV7670摄像头模块作为主控芯片进行图像数据采集处理.通过图像采集、灰度转换、二值化、图像腐蚀与膨胀和边界曲线提取等处理过程,在满足图像处理质量的前提下,提高边界图像处理的稳定性和实时性.该方法在图像处理过程中无需借助计算机实现,极大改善了户外型割草机器人的智能性,且工作效率高、响应快,为割草机器人的完全智能化设计提供了保障.%Moving robot vision is one of the key technologies of its boundary image recognition. This paper uses ARM series of STM32F103ZE and OV7670 camera modules as the master control chip for image data acquisition processing. On the premise of meeting the quality requirement of image processing, the stability of real time image processing is improved through acquisition of image,image transformation of gray level, corrosion and expansion of image and boundary curve extraction process. In the process of image processing without the computer implementation, this method is used to greatly improve the intelligence of outdoor moving robot.It is characteristic of high efficiency and fast response. The basis is provided for its intelligent design.