Open AccessA Fast Method for Image Matching and Registration Based on SIFT Algorithm and Image Pyramid
Author(s) -
Chunyu Guo,
Fangxiu Jia,
Wei Tang,
Pan Huang
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1449/1/012119
Subject(s) - scale invariant feature transform , pyramid (geometry) , artificial intelligence , feature (linguistics) , feature detection (computer vision) , image (mathematics) , computer vision , computer science , image registration , pattern recognition (psychology) , matching (statistics) , image processing , mathematics , linguistics , philosophy , statistics , geometry
In this paper, a fast image registration algorithm based on SIFT is proposed, which is slow in extracting feature points from high-resolution images. In this paper, the original image is transformed into an image pyramid to obtain a sub-sampled image with a lower resolution. At the same time, the Laplace image pyramid is obtained. Feature points of the sub-sampled image are extracted through SIFT algorithm, and then they are fused with the original image to obtain feature points of the original image. Experimental results show that the algorithm can quickly find the feature points and obviously improve the computational efficiency.