A local feature extraction method for UAV-based image registration based on virtual line descriptors
- PDF / 1,331,758 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 75 Downloads / 226 Views
ORIGINAL PAPER
A local feature extraction method for UAV-based image registration based on virtual line descriptors Lei Xing1 · Wujiao Dai1 Received: 8 April 2020 / Revised: 14 August 2020 / Accepted: 16 September 2020 © Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract Image local feature extraction is extensively utilized in the field of photogrammetry where the spatial distribution of features is important in high-quality image matching, particularly in high-resolution unmanned aerial vehicle (UAV) image registration. Presently, the spatial distribution problems are considered in some local feature extraction methods, though these methods are designed for point descriptors. Line descriptors are more robust to repetitive patterns compared to point descriptors and have attracted extensive attention in recent years. Hence, a feature extraction method is designed in this paper for line descriptors based on the K-connected virtual line descriptors matching method. Using the four measures, the quality of local features is quantified, and a regular gridding strategy based on the quality of local features is applied in the feature selection procedure. The proposed feature extraction method was successfully applied to match various simulated and real UAV-based images. Based on the experimental results using real images, it is indicated that two evaluation criteria, namely the spatial distribution quality of features and the number of correct matches, are improved to at least 12% and 15%, respectively, for verifying the capability of the proposed method to enhance matching performance. Keywords Image matching · Local feature extraction · Line descriptors · Spatial distribution
1 Introduction Image matching plays a vital role in computer vision or digital photogrammetry tasks such as structure from motion [1] and 3D model reconstruction [2]. Numerous studies have conducted on high-precision, fully automated image matching methods. In recent years, the research on invariant detectors has gradually emerged within rapid development, of which the SIFT (scale invariant feature transformation) algorithm and its variants are the most representatives [3, 4]. The feature points extracted by invariant detectors are known as local features. Because only local feature regions are used in feature descriptors, there will be a huge deal of ambiguity in initial image matching such as false matches, even with the help of ratio-test and crosscheck methods. Additionally, regions with repetitive patterns also cause difficulties in image matching. Therefore, for high-precision and robust UAV image regis-
B 1
Wujiao Dai [email protected] School of Geosciences and Info-Physics, Central South University, Changsha, China
tration, it is essential to achieve effective and robust image feature matching and outlier removal. The existing methods of outlier removal are in two kinds including methods based on geometric constraints and methods based on photometric constraints. In the first group, the local or global geometric constr
Data Loading...
