多視立體視覺三維重建技術於工程監測與變遷偵測應用

ENGINEERING MONITORING AND CHANGE DETECTION FOR MULTI-VIEW STEREO 3D RECONSTRUCTION TECHNOLOGY

張庭榮(Ting-Rong Chang) ； 李良輝(Liang-Hwei Lee)

《中國土木水利工程學刊》 31卷4期 (2019/06) Pp. 337-350

https://doi.org/10.6652/JoCICHE.201906_31(4).0005

摘要

多視立體視覺三維重建技術之發展，補足了傳統近景攝影測量技術之不足。如全自動的影像匹配技術，使相機自率定及外方位參數（四元數）推求得以更加快速方便；多視立體視覺立體匹配技術，可獲取大量三維點雲資料等。利用全自動之影像匹配技術，配合稠密點雲資訊將可有效描述三維重建本體，進而應用於整體及變遷偵測使用。本研究提出以多視立體視覺三維重建技術為主，提供簡單、廉價及精度高之工程監測解決方案，並以邊坡自動監測、港區結構體監測及三維變遷偵測為例，進行相關測試及驗證，以證明該方法用於工程監測之可行性。

關鍵字

多視立體視覺 ； 工程監測 ； 近景攝影 ； 三維重建 ； 點雲 ； 變遷偵測

並列摘要

The development of Structure from Motion multi-view stereo 3D reconstruction technology complements the shortcomings of traditional close-range Photogrammetry. Such as automatic image matching technology, the camera self-calibration and external orientation parameters (Quaternion) to inquire more quickly and easily; Multi-viewer stereo matching technology, access to a large number of 3D point cloud data. Using the fully automatic image matching technology, with the dense point cloud will be able to effectively describe the 3D reconstructed structure, and then applied to the overall and the use of change detection. This study proposes to multi-view stereo 3D reconstruction technology-based, to provide a simple, low-cost and high-precision monitoring of engineering solutions, and slope automatic monitoring, port monitoring of the structure and 3D change detection as an example, relevant test, and verification to prove the feasibility of this method for the monitoring project.

並列關鍵字

multi-view stereo ； engineering monitoring ； close-range photogrammetry ； 3D reconstruction ； point cloud ； 3D change detection

參考文獻

Hartley, R. I.(1997).Kruppa’s equations derived from the fundamental matrix.IEEE Transactions on Pattern Analysis and Machine Intelligence.19(2),133-135.

Hartley, R.,Zisserman, A.(2004).Multiple View Geometry in Computer Vision.Cambridge University Press.

Lowe, D. G.(2004).Distinctive image features from scaleinvariant keypoints.International Journal of Computer Vision (IJCV).60,91-110.

Furukawa, Y.,Ponce, J.(2010).Accurate, Dense and robust multiview stereopsis.IEEE Transactions on Pattern Analysis and Machine Intelligence.32,1362-1376.

Snavely, N.,Seitz, S. M.,Szeliski, R.(2008).Modeling the world from internet photo collections.International Journal of Computer Vision.80(2),189-210.