A Feature-Based Aerial Image Mapping Algorithm for UAV

DOI

10.2991/aer.k.201203.021How to use a DOI?

Keywords

UAV, Image Stitching, Feature Extraction, Feature Matching

Abstract

At present, the image stitching algorithm based on classical surf features is facing new challenges in the processing of UAV images. In order to improve the efficiency of aerial image stitching, a fast feature extraction and matching algorithm is proposed. At the feature extraction link, a local differential binary algorithm is proposed to describe the feature, which reduces the feature dimension compared with SURF descriptor while not reducing the feature differentiation. A local sensitive hash search algorithm is proposed to replace the kd tree search algorithm in feature matching, which improves the efficiency of nearest neighbor feature matching. The test results show that compared with the nearest neighbor matching algorithm based on SURF descriptor and kd tree search algorithm, the feature matching efficiency of this algorithm is obviously improved, and the matching accuracy is also improved. It is more suitable for feature-based UAV aerial image rapid mapping.

Copyright

© 2020, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Bin Liao
AU  - Hong Li
PY  - 2020
DA  - 2020/12/03
TI  - A Feature-Based Aerial Image Mapping Algorithm for UAV
BT  - Proceedings of the 2020 9th International Conference on Applied Science, Engineering and Technology (ICASET 2020)
PB  - Atlantis Press
SP  - 105
EP  - 108
SN  - 2352-5401
UR  - https://doi.org/10.2991/aer.k.201203.021
DO  - 10.2991/aer.k.201203.021
ID  - Liao2020
ER  -