3D indoor mapping using a RGB-D camera system
Corressponding author's email:
nthai@hcmute.edu.vnKeywords:
RGB-D camera system, 3D indoor mapping, SIFT, transformation matrix, point cloudsAbstract
This research proposes an application of a RGB Depth (RGB-D) camera system in 3D indoor mapping. In order to reconstruct the 3D model of an indoor space where the robot is located, the RGB-D camera system is installed in the robot frame to continuously capture the separated 2D image frames. All corresponding 2D points between the two consecutive image frames are firstly estimated using a Scan Invariant Feature Transform (SIFT) algorithm. Secondly, all pixel coordinates of these matching points are projected to the respective 3D space based on the depth information from the RGB-D camera at each pixel. The current outputs of this stage are the 3D points in the two successive point clouds that are applied to find out the transformation matrix. Finally, the estimated matrix is used to transform the second point cloud to the coordinating system of the first one. The result after repeating the above process with other consecutive pair of image frames and point clouds is the 3D model of the navigational space. The aimed method is relatively low cost configuration; furthermore, its accuracy is acceptable with the indoor robot applications.
Downloads: 0
References
J. Fuentes-Pacheco, Visual simultaneous localization and mapping: a survey, in Springer Science+Business Media Dordrecht, ed, pp. 55–81, 2015.
A. Elfes, Sonar-based real-world mapping and navigation, in IEEE Journal on Robotics and Automation vol. 3, ed, pp. 249-265, 1987.
C. P. O. Diaz and A. J. Alvares, A 2D-mapping strategy based on line segment extraction, in ANDESCON, 2010 IEEE, ed, pp. 1-6, 2010.
S. Ogawa, K. Watanabe, and K. Kobayashi, 2D mapping of a closed area by a range sensor, in SICE 2002. Proceedings of the 41st SICE Annual Conference vol. 2, ed, pp. 1329-1333 vol.2, 2002.
R. Z. M. Dr.Wael R. Abdulmajeed, Comparison Between 2D and 3D Mapping For Indoor Environments, in International Journal of Engineering Research and Technology vol. 2, ed, 2013.
J. L. Cras and J. Paxman, A modular hybrid SLAM for the 3D mapping of large scale environments, in Control Automation Robotics & Vision (ICARCV), 2012 12th International Conference on, ed, pp. 1036-1041, 2012.
L. Hung-Hsing, T. Ching-Chih, H. Ssu-Min, and C. Hsu-Yang, Automatic mapping for an indoor mobile robot assistant using RFID and laser scanner, in SICE, 2007 Annual Conference, ed, pp. 2102-2108, 2007.
D. Schleicher, L. M. Bergasa, R. Barea, E. Lopez, and M. Ocana, Real-Time Simultaneous Localization and Mapping Using a Wide-Angle Stereo Camera, in Distributed Intelligent Systems: Collective Intelligence and Its Applications, 2006. DIS 2006. IEEE Workshop on, ed, pp. 55-60, 2006.
C. Lim Chee, S. N. Basah, S. Yaacob, M. Y. Din, and Y. E. Juan, Accuracy and reliability of optimum distance for high performance Kinect Sensor, in Biomedical Engineering (ICoBE), 2015 2nd International Conference on, ed, pp. 1-7, 2015.
D. G. Lowe, Distinctive Image Features from Scale-Invariant Keypoints, in International Journal of Computer Vision vol. 60, ed, pp. 91-110, 2004.
M. Wei-wei, L. My-Ha, and J. Kang-Hyun, 3D reconstruction and measurement of indoor object using stereo camera, in Strategic Technology (IFOST), 2011 6th International Forum on vol. 2, ed, pp. 738-742, 2011.
X. Huang and M. Hu, 3D Reconstruction Based on Model Registration Using RANSAC-ICP Algorithm, in Transactions on Edutainment XI, Z. Pan, D. A. Cheok, W. Mueller, and M. Zhang, Eds., ed. Berlin, Heidelberg: Springer Berlin Heidelberg, pp. 46-51, 2015.
Downloads
Published
How to Cite
Issue
Section
Categories
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Copyright © JTE.
