A Dimension-Reduced SIFT Approach for Visual SLAM of Mobile Robots
The detecting and matching efficiency of scale-invariant feature transform (SIFT) method,which is widely used in simultaneous localization and mapping (SLAM),has received a lot of attention from researchers.Through local information extraction,principal component analysis (PCA) and a matching correction procedure,a dimension-reduced SIFT approach is introduced in this paper,which greatly cuts down the processing time cost and tremendously improves the matching accuracy.Based on the combination of the improved SIFT approach with an extended Kalman filter (EKF) method,simulations on localization and mapping of mobile robots are carried out.The simulation results show the efficiency of our approach.
SLAM SIFT dimension-reduced EKF mobile robots
WU Yuan-yuan HUANG Yi JIA Ying-min
The Seventh Research Division, Beihang University(BUAA), Beijing 100191, China
国内会议
河南焦作
英文
1-6
2014-07-26（万方平台首次上网日期，不代表论文的发表时间）