论文部分内容阅读
针对飞行机器人巡检的海量电力杆塔图像,以电力杆塔和架空电力线路巡检飞行机器人间的相对位置关系为先验信息,估计电力杆塔在飞行机器人巡检拍摄得到图像中的位置范围.然后利用这个位置范围作为可变型部件模型(deformable model part,DPM)检测区域的约束,从而精确快速定位电力杆塔的位置.实验结果表明:通过飞行机器人巡检获取的视频图像进行测试,验证了融合地理位置信息后的可变型部件模型的方法(GI-DPM)的有效性,提高了电力杆塔的检测速度.
Aiming at the image of massive power tower of the flying robot inspection, taking the relative position between the flying tower and overhead power line as the priori information, the position of the power tower in the image taken by the flying robot is estimated. Then, This location range is used as a constraint of the deformable model part (DPM) detection area to accurately and quickly locate the position of the power tower.The experimental results show that the video images obtained by the inspection of flying robots are tested and the fusion geographical location The effectiveness of the information-based mutable component model approach (GI-DPM) improves the detection speed of power towers.