Matching of line features with projective invariants

Djemaa Kachi; Xiaowei Tu

doi:10.1117/12.216405

11 August 1995 Matching of line features with projective invariants

Djemaa Kachi, Xiaowei Tu

Proceedings Volume 2573, Vision Geometry IV; (1995) https://doi.org/10.1117/12.216405
Event: SPIE's 1995 International Symposium on Optical Science, Engineering, and Instrumentation, 1995, San Diego, CA, United States

Abstract

This paper addresses the problem of line invariant features matching in a sequence of stereoscopic images of flat objects. The proposed approach is simple, but effective. It is essentially based on the computing of line features which are invariants under the geometrical transformation, such as projective or affine transformation. A line segment is represented by an invariant feature vector in a bi-dimensional parameter space. The line matching is achieved through two processes, the hypothesize and verification process. Each invariant feature vector is directly compared with the pre-registered feature vectors. A pair of lines which are represented by corresponding feature vectors must satisfy some pre-defined geometric constraints (relative angle and distance), in order to be considered matched. Our method is an efficient algorithm which is distinctly fast. Both computer generated and real data are included in experiments to show the stability of the computed features and the convergence speed of the matching system.

Citation Download Citation

Djemaa Kachi and Xiaowei Tu "Matching of line features with projective invariants", Proc. SPIE 2573, Vision Geometry IV, (11 August 1995); https://doi.org/10.1117/12.216405

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