Loop-closing By Using SIFT Features for Mobile Robots

Abstract

To resolve the problem of accumulative error in the real-time loop-closing of SLAM, SIFT features are used to match the visual appearance in the environment. The visually salient regions are selected from an image using Scale Saliency algorithm. Then, the SIFT features extracted from the salient regions in current image are used to match to the possible candidate features. The experiment results show the method is feasible.