Distance-Scale Restricted SIFT Matching Method for Complicated Content & Different Scales Image Pairs

Zhen Yu Wu; Hu Hong

doi:10.4028/www.scientific.net/AMM.526.292

Paper Titles

Simulation and Design of Control and Display Unit for Full Flight Simulator
p.273

Design of Smart Car Video Capture System Based on SAA7113
p.277

Analysis of Driver Visual Characteristics Based on FaceLAB
p.283

Application of a Programmable Partial Discharge Calibrator
p.288

Distance-Scale Restricted SIFT Matching Method for Complicated Content & Different Scales Image Pairs
p.292

A Data Center Energy-Saving System and Method Based on Virtual IT Resource Distribution Technology
p.297

The Research on Energy-Saving Strategy of WSN in 220KV Substation
p.303

Based on Image Processing Technology of Trajectory Monitoring Analysis Research
p.308

The Research on Intelligent Wireless Communications Robot Collaboration Control System
p.312

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vol. 526Distance-Scale Restricted SIFT Matching Method for...

Distance-Scale Restricted SIFT Matching Method for Complicated Content & Different Scales Image Pairs

Abstract:

Scale invariant feature transform (SIFT) matching performance decreases greatly when images are in different scales with complicated content and wide-baseline. In this paper, we address this problem, and propose a simple method to improve SIFT matching. The proposed method restricts the matching searching area into much smaller and more likely region to improve matching performance. Experiments shows that the proposed method has saved up to 90% matching time and increased up to 4% in the accuracy, compared with SIFT and previously solutions which only improve the matching accuracy.

You might also be interested in these eBooks

Mechanical Engineering and Instrumentation

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volume 526)

Pages:

292-296

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.526.292

Citation:

Cite this paper

Online since:

February 2014

Authors:

Zhen Yu Wu*, Hu Hong

Keywords:

Distance-Scale, Image Retrieval, Keypoints, Matching, SIFT

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

* - Corresponding Author

References

[1] Lowe, D.: Distinctive image features from scale invariant keypoints, Int. J. Comput. Vis., (2004), pp.91-110.

Google Scholar

[2] Lowe, D. G. Object recognition from local scale-invariant features, Int. Conf. on Computer Vision, (1999）, pp.1150-1157.

Google Scholar

[3] L. Li, X. Long: LSIFT-an improved SIFT algorithm with a new matching method, in Proc. ICCASM, Taiyuan, China, Oct. (2010), pp.643-646.

DOI: 10.1109/iccasm.2010.5620607

Google Scholar

[4] Yi, Z., Zhiguo, C., and Yang, X.: Multi-spectral remote image registration based on SIFT, Electron. Lett., (2008), 44, (2), pp.107-108.

DOI: 10.1049/el:20082477

Google Scholar

[5] Alhwarin, F., Wang, C., Ristic-Durrant, D., and Graser, A.: Improved SIFT-features matching for object recognition, Visions of Computer Science – BCS International Academic Conf., Imperial College, London, UK, Sept. (2008), pp.179-190.

DOI: 10.14236/ewic/vocs2008.16

Google Scholar

[6] Bastanlar, Y., Temizel A. and Yardmc Y.: Improved SIFT matching for image pairs with scale difference, Electron. Lett., (2010), 46, (5), pp.346-348.

DOI: 10.1049/el.2010.2548

Google Scholar