公开(公告)号：US20170004345A1

公开(公告)日：2017-01-05

申请号：US15196425

申请日：2016-06-29

Applicant: TOPCON CORPORATION

Inventor： Takeshi SASAKI ,  Tetsuji ANAI ,  Hitoshi OTANI ,  Nobuo KOCHI

IPC: G06K9/00 ,  G06T7/00

CPC classification number: G06K9/0063 ,  G01C11/06 ,  G06T17/05 ,  G06T2207/10028 ,  G06T2207/10032

Abstract: The efficiency of the work for identifying reference points that are included in photographed images is improved. Targets are located and are photographed from short distances, and the positions of the targets are measured. A 3D reference point model is generated by using the measured positions of the targets as apexes. Then, the positions of the targets that are detected from images taken by a UAV from the air are calculated from a three-dimensional model that are generated by the principle of the stereoscopic three-dimensional measurement, whereby a 3D relative model constituted of a TIN is obtained. After a matching relationship between the 3D reference point model and the 3D relative model is identified, the positions of the targets in the 3D relative model are estimated based on the identified matching relationship, and the positions of the reference points in the images taken from the UAV are estimated.

Abstract translation: 提高了拍摄图像中包含的识别参考点的工作效率。 定位目标，并从短距离拍摄，并对目标的位置进行测量。 通过使用目标的测量位置作为顶点来生成3D参考点模型。 然后，从由立体三维测量的原理产生的三维模型中计算从UAV从空中拍摄的图像中检测到的目标的位置，由此由TIN构成的3D相对模型 获得。 在3D参考点模型和3D相对模型之间的匹配关系被识别之后，基于所识别的匹配关系来估计3D相对模型中的目标的位置，并且从图像中获取的参考点的位置 无人机估计。