无人机管线巡检中的测试桩识别

在无人机GPS信号丢失的情况下，测试桩的视觉辅助识别精度是影响油气管线自动巡检工作的关键因素。针对测试桩自动识别的精度问题，在分析测试桩及其周围地物背景目标特性的基础上，先用深度学习算法判断出测试桩被周围地物背景遮挡的情况。对于被遮挡的测试桩，采用不显著目标相对定位算法检测出测试桩的具体位置。最后通过现场采集的数据实验验证了文中算法的有效性。

Test-pile Detection in Pipeline Inspection by UAV

In the case of GPS signal loss of the unmanned aerial vehicle (UAV), the auxiliary visual recognition accuracy of the test pile is a key factor affecting the automatic inspection of oil and gas pipelines. Aiming at the accuracy problem of automatic identification of test piles, based on the analysis of the background and target characteristics of the test piles and surrounding objects, a deep learning algorithm was used to determine whether the test piles were obscured by the surrounding objects. For obstructed test piles, the relative location algorithm of the insignificant target was used to detect the specific position of the test piles. Finally, the validity of the algorithm in this paper is verified by actual test scenarios experiments.