Research on target attitude estimation method based on image

Aiming at the real time and accuracy of door handle pose estimation, this paper improved ORB matching algorithm and combined with least square plane fitting algorithm. In order to improve the real-time performance of the system to obtain the location information of the door handle, the ORB matching algorithm was used to reduce the time complexity of the algorithm itself and the polar constraint algorithm was added to reduce the search dimension to straight line search to reduce the matching time and reduce the mismatching. Finally, the least square plane fitting algorithm is used to obtain the spatial attitude information of the door handle under the camera plane. The accuracy and real-time performance of the improved ORB matching algorithm and the accuracy of the door handle pose are verified by experiments. The proposed method can meet the requirements in practical application.