Open AccessBuilding Recognition System Based on Improved SIFT Algorithm and Positioning Information
Author(s) -
Tao Liu,
Feng Jiang,
Yu Gao
Publication year - 2021
Publication title -
frontiers in signal processing
Language(s) - English
Resource type - Journals
eISSN - 2521-7380
pISSN - 2521-7372
DOI - 10.22606/fsp.2021.53002
Subject(s) - scale invariant feature transform , computer science , global positioning system , artificial intelligence , computer vision , order (exchange) , positioning system , algorithm , image (mathematics) , data mining , pattern recognition (psychology) , engineering , telecommunications , structural engineering , finance , node (physics) , economics
In order to solve the general problem, that is, the accurate recognition rate is low in a small extent or when the image resources are few and scattered. This article puts forward a building recognition system that combines GPS positioning information with an improved SIFT algorithm, and adds a pre-processing mechanism to predict the possibility of the building existence in the system, which further reduces mismatch and improves response speed. The final verification shows that this research is actually effective.