Open AccessImproved SIFT algorithm based on image filtering
Author(s) -
Mingyu Qiao,
Xiao Liang,
Minjie Chen
Publication year - 2021
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 - 10.1088/1742-6596/1848/1/012069
Subject(s) - scale invariant feature transform , enhanced data rates for gsm evolution , matching (statistics) , feature (linguistics) , feature extraction , artificial intelligence , pattern recognition (psychology) , image (mathematics) , computer science , edge detection , computer vision , feature detection (computer vision) , algorithm , mathematics , image processing , statistics , linguistics , philosophy
In this paper, SIFT feature extraction algorithm was optimized through image filtering, so as to highlight the role of stable edge corner and improve the efficiency of stable edge corner collection. Afterwards, the experimental results were compared and verified by the FLANN feature matching method. Experimental results show that the improved SIFT feature extraction algorithm using image filtering can improve the extraction effect of feature points with stable edge response, while suppressing the extraction of feature points with unstable edge response, thus improving the accuracy of matching.