摘要
随着DSP处理能力的不断增强以及图像处理算法的日益完善,开发基于机器视觉的嵌入式实时系统成为目前图像处理领域一项重要分支。本文提出了一种基于DSP(TMS320DM642)的石油管道焊缝检测机器人系统,用于完成高性能的石油管道焊缝的实时自动定位和检测任务;介绍了机器人系统的工作原理,硬件系统原理;阐述了系统的焊缝识别算法和基于RF5参考架构的软件框架设计方案。经验证,该识别算法可靠、稳定、实时性好,能满足石油管道焊缝检测机器人系统对焊缝快速准确定位的要求。
With the ever strengthening processing ability of DSP and the ever perfecting of image processing algorithms, the development of real-time embedded system based on machine vision becomes an important branch of image processing. A design of petroleum pipe welding line detecting robot system based on DSP （TMS320DM642） was put forward to finish high quality tasks of real-time automatic positioning and detecting, working theory of this robot system and hardware system were introduced, and welding line recognition algorithm of the system and software design framework based on RF5 reference frame was illustrated. Experiments show that the recognition algorithm is reliable, stable and has real-time performance, thus can satisfy the requirements of the petroleum pipe welding line detecting robot system for fast and accurate positioning.
出处
《电子测量技术》
2007年第1期144-146,共3页
Electronic Measurement Technology
关键词
DSP
嵌入式
图像处理
管道焊缝检测
DSP
embedded
image processing
pipe welding line detecting
