8 February 2017 Error-detective one-dimensional mapping
Author Affiliations +
Proceedings Volume 10225, Eighth International Conference on Graphic and Image Processing (ICGIP 2016); 102250O (2017) https://doi.org/10.1117/12.2267116
Event: Eighth International Conference on Graphic and Image Processing, 2016, Tokyo, Japan
Abstract
The 1-D mapping is an intensity-based method used to estimate a projective transformation between two images. However, it lacks an intensity-invariant criterion for deciding whether two images can be aligned or not. The paper proposes a novel decision criterion and, thus, develops an error-detective 1-D mapping method. First, a multiple 1-D mapping scheme is devised for yielding redundant estimates of an image transformation. Then, a voting scheme is proposed for verifying these multiple estimates, in which at least one estimate without receiving all the votes is taken as a decision criterion for false-match rejection. Based on the decision criterion, an error-detective 1-D mapping algorithm is also constructed. Finally, the proposed algorithm is evaluated in registering real image pairs with a large range of projective transformations.
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Yun Zhang, Yun Zhang, Shihao Zhou, Shihao Zhou, } "Error-detective one-dimensional mapping", Proc. SPIE 10225, Eighth International Conference on Graphic and Image Processing (ICGIP 2016), 102250O (8 February 2017); doi: 10.1117/12.2267116; https://doi.org/10.1117/12.2267116
PROCEEDINGS
6 PAGES


SHARE
Back to Top