Fusing Dense Features and Pose Consistency: A Regression Method for Attitude Measurement of Aircraft Landing

The aircraft pose is the important measurement indicator related to a safe landing. The existing pose estimation methods are limited by the extraction accuracy of salient sparse points or line features, and it is hard to obtain higher accuracy and robustness, especially at long distances. In this article, we implement attitude measurement using pose estimation. The aircraft pose estimation method based on dense features and pose consistency (PC) is proposed, which can infer the aircraft pose relative to the camera from the RGB image directly. This method uses the encoder-decoder network to predict the dense 3-D coordinates, the surface regions (SRs), and visible segmentation (VS). Guided by the dense 2D-3D correspondences (DCs) and SRs, the aircraft pose is estimated using the pose regression module (PRM). To improve the robustness of decoupled pose estimation relative to object detection, this article proposes a dual-channel regression framework connected with the PC constraint, which enables mutual supervision between the different dynamic zoomed-in views (DZIs). The experimental results on the aircraft dataset show the superiority of our aircraft pose estimation method significantly. The PC constraints further improve the prediction performance on the aircraft dataset, LINEMOD dataset, and YCB-V dataset. The pose estimation method can be used for real-time measurement of aircraft attitude directly with the mean error of 0.377°, the rms error of 0.491°, and the rate of 59 frames/s.