Automatic segmentation algorithm for dense steel structure point clouds

Yanping Liu; Weiping Zhang; Dongfeng Jia

doi:10.1117/12.2591242

28 February 2021 Automatic segmentation algorithm for dense steel structure point clouds

Yanping Liu, Weiping Zhang, Dongfeng Jia

Proceedings Volume 11781, 4th Optics Young Scientist Summit (OYSS 2020); 1178107 (2021) https://doi.org/10.1117/12.2591242
Event: Optics Frontier: Optics Young Scientist Summit, 2020, Ningbo, China

Abstract

For the problem of dense distribution of steel structures and low efficiency of interactive point clouds extraction, an automatic segmentation algorithm of dense steel structure point clouds is proposed in the paper. Firstly, point clouds are divided into several sub blocks by Octree, meanwhile, its spatial topological neighborhood is established in the process. According to the structural characteristics of angle steel components, the Random Sample Consensus algorithm based on additional normal vector constraints is implemented to search large area plane in the sub block, and then, Euclidean distance clustering and area growth algorithm with additional smooth constraints are put forward to segment the secondary data until steel structure point clouds is clearly segmented. The validity and accuracy of the algorithm are verified by real transmission tower point clouds. The experimental results show that the automatic segmentation algorithm proposed in the paper can segment the steel structure point clouds quickly and accurately, and has high application value in future.

Citation Download Citation

Yanping Liu, Weiping Zhang, and Dongfeng Jia "Automatic segmentation algorithm for dense steel structure point clouds", Proc. SPIE 11781, 4th Optics Young Scientist Summit (OYSS 2020), 1178107 (28 February 2021); https://doi.org/10.1117/12.2591242

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available