Comparison of plane extraction performance using laser scanner and Kinect

Abstract

This paper compares the performance of laser scanner and Kinect when extracting the plane from 3D depth map measured by both sensors. By tilting laser scanner, 3D depth map around robot is generated in this paper and another 3D depth map using Kinect is also generated. To compare the 3D depth map of laser scanner and Kinect, staircases are measured by both sensors and the accuracy of the extracted planes of the staircases is examined. Although laser scanner needs more time to generate 3D depth map, laser scanner-based surface extraction shows more accurate result.