Robust indoor scene recognition based on 3D laser scanning and Bearing Angle image

Abstract

Robust scene recognition serves as an essential task for robots to work within a complex dynamic environment. Considering vision device's limited adaptability in the dark environment, a 3D-laser-based scene recognition approach that extracts and matches SIFT features from Bearing Angle images is proposed, which makes it possible to make full use of both global metric information and local scale-invariant features. This approach can not only cope with irregular disturbances of dynamic objects, but also tackle obvious changes of observation location robustly in a semi-structured environment. An large-scale indoor environment with more than 30 offices is selected as the real-world scenes to test the performance of the proposed approach.