Abstract
Terrain classification allows a mobile robot to create an annotated map of its local environment from the three-dimensional (3D) and two-dimensional (2D) datasets collected by its array of sensors, including a GPS receiver, gyroscope, video camera, and range sensor. However, parts of objects that are outside the measurement range of the range sensor will not be detected. To overcome this problem, this paper describes an edge estimation method for complete scene recovery and complete terrain reconstruction. Here, the Gibbs-Markov random field is used to segment the ground from 2D videos and 3D point clouds. Further, a masking method is proposed to classify buildings and trees in a terrain mesh.
Highlights
Object segmentation and classification are widely researched topics in surveying, mapping, and autonomous navigation by mobile robots [1,2]
The multiple sensors mounted on such robots collect terrain information only in the form of three-dimensional (3D) point clouds and two-dimensional (2D) images [4]
We propose the Gibbs-Markov random field (MRF) method that detects the boundary pixels between objects and background in order to recover the missing tops of objects
Summary
Object segmentation and classification are widely researched topics in surveying, mapping, and autonomous navigation by mobile robots [1,2]. These techniques allow a robot to navigate through and interact with its environment by providing quickly accessible and accurate information regarding the surrounding terrain [3]. Especially ground-based autonomous robots, detect surrounding terrain information, some parts of objects are outside the measurement of range sensors. The present paper is organized as follows: in Section 2, we discuss related work on multisensor integration, interpolation, ground segmentation, and object classification in real-world applications. In. Section 4, we analyze the results of the proposed ground segmentation, height estimation, and object classification methods.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have