基于吊钩动作智能跟踪的塔机图像监控

为了对吊钩及重物及时有效地捕捉监控，开发了可智能跟踪塔机吊钩及重物的图像监控系统。首先根据塔机吊绳特点，基于分步概率Hough变换，通过优化设计累加器数组、制定合适的规则进行线段搜索和筛选结果等手段，设计了适用于嵌入式系统的塔机吊绳识别算法；进而通过吊绳末端坐标得到监控图像中的吊钩位置，然后控制云台实现了摄像机对塔机吊钩动作的智能跟踪，从而保证了吊钩及重物始终处于监测视野的中央。实验结果表明该算法可有效识别塔机吊绳，并且运算效率较高，适用于运算能力有限的嵌入式系统。

A Video Surveillance System for Tower Cranes Using Automatic Tracking of Hook Movements

This paper presents a video surveillance system that tracks hook movements automatically ：first based on the progressive probabilistic Hough transform, by designing an optimized accumulator, establishing proper rules of hook＇s position in ＇the searching line segments and filtering results, a sling-detection algorithm is proposed thus the input image canbe acquired by determining the coordinates of the end point of the sling. Then a PTZ camera is used to track the hook＇s movements. Experimental results show that the algorithm is effective in detecting slings and, because of the high efficiency of the algorithm, feasible to operate in embedded systems.