Abstract For SIFT(Scale Invariant Feature Transform) has poor real-time and low match rate problem in large-scale image registration, a improved registration algorithm based on SIFT algorithm is proposed in this paper, which down-sample the large-scale image with Cubic interpolation algorithm, and under the restriction that the minimum size of down-sampling image should be meet image registration requirement. We can calculate the transformation matrix of down-sampling image with SIFT algorithm, and then we can get transformation matrix of original large-scale image through the relationship of transformation matrix between down-sampling image and original image, so the registration of large-scale image can be got more rapidly. The experiment results show the effectiveness and feasibility of the proposed method.
Research on Novel Optimization SIFT Algorithm Based Fast Mosaic Method
2014-01-01
10 pages
Article/Chapter (Book)
Electronic Resource
English
Research on Novel Optimization SIFT Algorithm Based Fast Mosaic Method
British Library Conference Proceedings | 2014
|Dynamic image mosaic via SIFT and dynamic programming
British Library Online Contents | 2014
|British Library Conference Proceedings | 2006
|A research on Quick-SIFT and ghosting elimination technique for UAV image mosaic
British Library Conference Proceedings | 2020
|The Research of Video Tracking Based on Improved SIFT Algorithm
British Library Conference Proceedings | 2016
|