Main Image
Volume 299 - The 7th International Conference on Computer Engineering and Networks (CENet2017) - Session I - Machine Learning
Region-based Pre-selection of Overlapping Area for the SIFT Algorithm
Y. Zhou,* H. Wang, M. Jiang
*corresponding author
Full text: pdf
Pre-published on: 2017-07-17 11:56:25
Published on: 2017-09-06 14:01:04
Abstract
This paper presents a novel method of determining the overlapping area in advance to accelerate the SIFT (Scale Invariant Feature Transformation) algorithm. The SIFT algorithm can achieve high accuracy in image matching, however, it can be time-consuming when processing large images. The proposed method in this paper pre-selects the overlapping area of input images to speed up the feature extraction of SIFT. This improvement of SIFT can greatly reduce the
processing time without affecting the accuracy.
Open Access
Creative Commons LicenseCopyright owned by the author(s) under the term of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.