23 January 2017 Efficient content-based low-altitude images correlated network and strips reconstruction
Author Affiliations +
Proceedings Volume 10322, Seventh International Conference on Electronics and Information Engineering; 103221A (2017) https://doi.org/10.1117/12.2265247
Event: Seventh International Conference on Electronics and Information Engineering, 2016, Nanjing, China
Abstract
The manual intervention method is widely used to reconstruct strips for further aerial triangulation in low-altitude photogrammetry. Clearly the method for fully automatic photogrammetric data processing is not an expected way. In this paper, we explore a content-based approach without manual intervention or external information for strips reconstruction. Feature descriptors in the local spatial patterns are extracted by SIFT to construct vocabulary tree, in which these features are encoded in terms of TF-IDF numerical statistical algorithm to generate new representation for each low-altitude image. Then images correlated network is reconstructed by similarity measure, image matching and geometric graph theory. Finally, strips are reconstructed automatically by tracing straight lines and growing adjacent images gradually. Experimental results show that the proposed approach is highly effective in automatically rearranging strips of lowaltitude images and can provide rough relative orientation for further aerial triangulation.
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Haiqing He, Haiqing He, Qi You, Qi You, Xiaoyong Chen, Xiaoyong Chen, } "Efficient content-based low-altitude images correlated network and strips reconstruction", Proc. SPIE 10322, Seventh International Conference on Electronics and Information Engineering, 103221A (23 January 2017); doi: 10.1117/12.2265247; https://doi.org/10.1117/12.2265247
PROCEEDINGS
7 PAGES


SHARE
Back to Top