Large-Scale Aircraft Pose Estimation System Based on Depth Cameras

In the fields of wind tunnel measurement and aerospace, real-time pose information aircraft is an important index. this paper, we propose a large-scale estimation system, in which depth cameras are used to scan entire model multiple directions. Using principal component analysis (PCA) featuring vectors as target coordinate system through transformation matrix for point cloud calibration aircraft, merge complete with cloud. An intrinsic shape signature (ISS) key extraction histograms orientations (SHOT) feature description form descriptors. The scale clouds reduced, coarse registration performed by matching random sample consensus (RANSAC) mismatching. robustness algorithm improved, initial achieved precise clouds. experimental results demonstrate that proposed can achieve angle accuracy 0.05°.