A novel feature matching algorithm based on shape-context between optical and infrared image

Zenglu Wan; Zheng Li; Jianliang Weng; Hongxia Gao

doi:10.1117/12.832973

30 October 2009 A novel feature matching algorithm based on shape-context between optical and infrared image

Zenglu Wan, Zheng Li, Jianliang Weng, Hongxia Gao

Proceedings Volume 7495, MIPPR 2009: Automatic Target Recognition and Image Analysis; 74953C (2009) https://doi.org/10.1117/12.832973
Event: Sixth International Symposium on Multispectral Image Processing and Pattern Recognition, 2009, Yichang, China

Abstract

Different-source image matching is still a difficult problem in computer vision. In this paper, we propose a novel algorithm based on edge information, shape context descriptor and graph matching to make a good performance in the different-source image matching field, especially optical and infrared image. First, as the edge information is a relatively stable in these two kinds of images, we detect the feature point based on the edge information which can reflect the construction information. Then shape context descriptor, which is rotation invariant, scale invariant and translation invariant, is adapted to describe the feature point. Finally, the graph matching algorithm is used for the image matching to eliminate the false matching point. From experimental results, our algorithm is efficient in different-source image matching.

Citation Download Citation

Zenglu Wan, Zheng Li, Jianliang Weng, and Hongxia Gao "A novel feature matching algorithm based on shape-context between optical and infrared image", Proc. SPIE 7495, MIPPR 2009: Automatic Target Recognition and Image Analysis, 74953C (30 October 2009); https://doi.org/10.1117/12.832973

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available