Assessing Keypoint Detection Effectiveness of SIFT Implementations on Restricted Image Datasets
Keywords:
SIFT, keypoint detection, feature extraction, restricted image datasetsAbstract
The Scale-Invariant Feature Transform (SIFT) algorithm, introduced by Lowe [1, 2], is a cornerstone in computer vision for robust feature detection and description. Its ability to extract distinctive keypoints invariant to scale, rotation, and illumination changes has made it indispensable for tasks such as object recognition, image stitching, and 3D reconstruction. While SIFT's theoretical robustness is well-established, practical implementations can vary in their performance, particularly when applied to small-scale image datasets. This article presents a comparative analysis of keypoint detection performance across different SIFT implementations, specifically focusing on their efficacy and efficiency on limited image collections. We evaluate metrics such as the number of detected keypoints, their distribution, and repeatability under various image transformations. Our findings highlight the nuances and trade-offs inherent in different SIFT library choices, providing valuable insights for researchers and practitioners working with constrained computational resources or specialized datasets.
References
D. G. Lowe, “Object recognition from local scale-invariant features,” Proceedings of the Seventh IEEE International Conference on Computer Vision, pp. 1150–1157 vol.2, 1999, doi: 10.1109/ICCV.1999.790410.
D. G. Lowe, “Distinctive image features from scale invariant keypoints,” International Journal of Computer Vision, vol. 60, pp. 91–11020042, 2004, doi: 10.1023/B:VISI.0000029664.99615.94.
Z. Wang, Y. Liu, J. Zhang, C. Fan, and H. Zhang, “Interference image registration combined by enhanced scale-invariant feature transform characteristics and correlation coefficient,” J Appl Remote Sens, vol. 16, no. 2, p. 26508, 2022.
S. Joseph, I. Hipiny, H. Ujir, S. F. S. Juan, and J. L. Minoi, “Performance evaluation of SIFT against common image deformations on iban plaited mat motif images,” Indonesian Journal of Electrical Engineering and Computer Science, vol. 23, no. 3, pp. 1470–1477, Sep. 2021, doi: 10.11591/ijeecs.v23.i3.pp1470-1477.
P. Selvaraj and M. Karuppiah, “Enhanced copy-paste forgery detection in digital images using scale-invariant feature transform,” IET Image Process, vol. 14, no. 3, pp. 462–471, Feb. 2020, doi: 10.1049/iet-ipr.2019.0842.
M. Kasiselvanathan, V. Sangeetha, and A. Kalaiselvi, “Palm pattern recognition using scale invariant feature transform,” International Journal of Intelligence and Sustainable Computing, vol. 1, no. 1, pp. 44–52, 2020.
R. Hess, “An open-source siftlibrary,” in Proceedings of the 18th ACM international conference on Multimedia, 2010, pp. 1493–1496.
I. Rey Otero and M. Delbracio, “Anatomy of the SIFT Method,” Image Processing On Line, vol. 4, pp. 370–396, Dec. 2014, doi: 10.5201/ipol.2014.82.
M. Y. Yin, F. Guan, P. Ding, and Z. F. Liu, “Implementation of image matching algorithm based on SIFT features,” in Applied Mechanics and Materials, Trans Tech Publications Ltd, 2014, pp. 3181–3184. doi: 10.4028/www.scientific.net/AMM.602-605.3181.
E. Yang, F. Chen, M. Wang, H. Cheng, and R. Liu, “Local Property of Depth Information in 3D Images and Its Application in Feature Matching,” Mathematics, vol. 11, no. 5, Mar. 2023, doi: 10.3390/math11051154.
Y. Gu, H. Wang, Y. Bie, R. Yang, and Y. Li, “Research on Image Registration of Different Wavebands Based on SIFT Algorithm,” in 2022 3rd China International SAR Symposium (CISS), 2022, pp. 1–5.
D. G. Lowe, “The SIFT Keypoint Detector,” 2005 [Online]. Available: https://www.cs.ubc.ca/~lowe/keypoints/.
R. Hess, “OpenSIFT: An Open-Source SIFT Library,” 2012 [Online]. Available: https://robwhess.github.io/opensift/.
A. Vedaldi and B. Fulkerson, “VLFeat - An open and portable library of computer vision algorithms,” Proceedings of the international conference on Multimedia - MM ’10, p. 1469, 2010, doi: 10.1145/1873951.1874249.
A. Krizhevsky, “Learning Multiple Layers of Features from Tiny Images,” 2009.
L. N. Darlow, E. J. Crowley, A. Antoniou, and A. J. Storkey, “CINIC-10 Is Not ImageNet or CIFAR-10.” 2018 [Online]. Available: https://datashare.is.ed.ac.uk/
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2023 Dr. Nora M. Al-Farhan
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors retain full copyright of their published work. By submitting to an ERDAST journal and upon acceptance of the article, the author(s) agree to grant the journal a non-exclusive license to publish, reproduce, distribute, and archive the article. All articles are published under the terms of the Creative Commons Attribution 4.0 International License (CC BY 4.0).
Under this license: Others may copy, distribute, display, remix, adapt, and build upon the work, even commercially, As long as proper credit is given to the original author(s) and source, A link to the license is provided, an, Any changes are clearly indicated.
???? License link: https://creativecommons.org/licenses/by/4.0/
