Paper
22 August 2015 Fast image mosaic algorithm based on the improved Harris-SIFT algorithm
Zetao Jiang, Min Liu
Author Affiliations +
Proceedings Volume 9656, International Symposium on Photonics and Optoelectronics 2015; 96560G (2015) https://doi.org/10.1117/12.2195450
Event: International Symposium on Photonics and Optics, 2015, Shanghai, China
Abstract
This paper proposes a fast image mosaic algorithm based on the improved Harris-SIFT algorithm, according to such problems as more memory consumption, greater redundancy quantity of feature points, slower operation speed, and so on, resulting from using the SIFT algorithm in the image matching stage of the image mosaic process. Firstly in the matching stage of the algorithm, the corner point is extracted by using the multi-scale Harris, feature descriptor is constructed by the 88-dimensional vector based on the SIFT feature, the coarse matching is carried out by the nearest neighbor matching method, and then the precise matching point pair and image transformation matrix are obtained by the RANSAC method. The seamless mosaic can be achieved by using the weighted average image fusion. The experimental results show that this algorithm can not only achieve precise seamless mosaic but also improve operation efficiency, compared with the traditional algorithm.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Zetao Jiang and Min Liu "Fast image mosaic algorithm based on the improved Harris-SIFT algorithm", Proc. SPIE 9656, International Symposium on Photonics and Optoelectronics 2015, 96560G (22 August 2015); https://doi.org/10.1117/12.2195450
Lens.org Logo
CITATIONS
Cited by 1 scholarly publication.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Image fusion

Image processing

Feature extraction

Image registration

Detection and tracking algorithms

Image resolution

Image compression

RELATED CONTENT


Back to Top