论文部分内容阅读
针对带式输送机输送带纵向撕裂故障,提出了一种基于ARM9的嵌入式图像处理技术进行输送带撕裂检测的新方法。利用Linux操作系统下线程互斥的方式,使图像采集和图像识别分别进行而互不干扰,而且在处理线程中直接将正常输送带图像剔除,从而不会发生内存容量增大的现象。此输送带纵向撕裂检测系统在唐山开滦东欢坨煤矿取得了良好的试验效果。与使用PC机相比,图像处理的效率和实时性都得到了很大提高,且大幅简化了系统结构,易于实现防爆,具有速度快、功耗低、体积小等特点。
Aiming at the longitudinal tearing fault of belt conveyor belt, a new method of belt tearing detection based on ARM9 embedded image processing technology was proposed. Using the mutual exclusion of threads under the Linux operating system, the image acquisition and image recognition are carried out without disturbing each other, and the image of the normal conveyor belt is directly removed from the processing thread so that the memory capacity will not increase. Longitudinal tear of the conveyor belt detection system in Tangshan Kailuan Dong Huan Tuo coal mine has achieved good results. Compared with the use of PC, image processing efficiency and real-time have been greatly improved, and greatly simplifies the system structure, easy to achieve explosion-proof, with high speed, low power consumption, small size and so on.