Translator Disclaimer
Poster + Paper
13 October 2020 Missing area reconstruction in 3D scene from multi-view satellite images for surveillance applications
Author Affiliations +
Conference Poster
Abstract
Automatic 3-D recovery from multiview satellite images can be extremely useful for information extraction for the surveillance application. For 3-D scene reconstruction, one approach is to employ multiple cameras for creating a multiview image with the aim to make interactive free-viewpoint selection possible in 3-D data. In most cases, such a 3-D scene contains missing holes on depth maps that appear during the synthesis from multi-views. This paper presents an automated pipeline for processing multi-view satellite images to 3-D digital surface models. The proposed approach uses the modified exemplar-based technique. We propose an algorithm using the concepts of a sparse representation of quaternions, which use a new gradient to calculate the priority function by integrating the structure of quaternions with LPA-ICI (local polynomial approximation - the intersection of confidence intervals) and the saliency map. Moreover, the color information incorporates into the optimization criteria to obtain sharp inpainting results. For this purpose, we use the Hamiltonian quaternion framework. Compared with state-of-the-art techniques, the proposed algorithm provides plausible restoration of the depth map from multi-view satellite images, which makes them a promising tool for surveillance applications.
© (2020) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
V. Voronin, N. Gapon, M. Zhdanova, E. Semenishchev, Y. Cen, and A. Zelensky "Missing area reconstruction in 3D scene from multi-view satellite images for surveillance applications", Proc. SPIE 11542, Counterterrorism, Crime Fighting, Forensics, and Surveillance Technologies IV, 115420P (13 October 2020); https://doi.org/10.1117/12.2574208
PROCEEDINGS
7 PAGES + POSTER

SHARE
Advertisement
Advertisement
Back to Top