{"title":"Visual Search Based Indoor Localization in Low Light via RGB-D Camera","authors":"Yali Zheng, Peipei Luo, Shinan Chen, Jiasheng Hao, Hong Cheng","volume":123,"journal":"International Journal of Computer and Information Engineering","pagesStart":403,"pagesEnd":407,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/10006850","abstract":"Most of traditional visual indoor navigation algorithms<br \/>\r\nand methods only consider the localization in ordinary daytime, while<br \/>\r\nwe focus on the indoor re-localization in low light in the paper. As<br \/>\r\nRGB images are degraded in low light, less discriminative infrared<br \/>\r\nand depth image pairs are taken, as the input, by RGB-D cameras, the<br \/>\r\nmost similar candidates, as the output, are searched from databases<br \/>\r\nwhich is built in the bag-of-word framework. Epipolar constraints can<br \/>\r\nbe used to relocalize the query infrared and depth image sequence.<br \/>\r\nWe evaluate our method in two datasets captured by Kinect2. The<br \/>\r\nresults demonstrate very promising re-localization results for indoor<br \/>\r\nnavigation system in low light environments.","references":"[1] Lowry, S., Sunderhauf, N., Newman, P., Leonard, J. J., Cox, D., Corke, P.,\r\nMilford, M. J., Visual Place Recognition: A Survey, IEEE Transactions\r\non Robotics, 2015, 31(1): 1\u201319.\r\n[2] Williams B., Cummins M., Neira J., Newman P., Reid I., Tardos J. D.,\r\nA comparison of loop closing techniques in monocular SLAM, Robotics\r\nand Autonomous Systems, vol. 57, no. 12, pp. 1188C1197, 2009\r\n[3] Lee D, Kim H, Myung H., 2D image feature-based real-time RGB-D 3D\r\nSLAM, Robot Intelligence Technology and Applications, 2012: 485-492.\r\n[4] Mur-Artal R, Montiel J. M. M., Tardos J D., ORB-SLAM: a versatile\r\nand accurate monocular SLAM system, IEEE Transactions on Robotics,\r\n2015, 31(5): 1147-1163.\r\n[5] Endres, F., Hess, J., Engelhard, N., Sturm, J., Cremers, D., Burgard, W.,\r\nAn Evaluation of the RGB-D SLAM System, ICRA, 2012.\r\n[6] Engel J, Schops T, Cremers D., LSD-SLAM: Large-scale direct\r\nmonocular SLAM, ECCV, 2014.\r\n[7] Salas-Moreno, R,, Newcombe, R., Strasdat, H., et al., Slam++:\r\nSimultaneous localisation and mapping at the level of objects, CVPR,\r\n2013.\r\n[8] Labbe M, Michaud F., Appearance-based loop closure detection for online\r\nlarge-scale and long-term operation, IEEE Transactions on Robotics,\r\n2013, 29(3): 734-745.\r\n[9] Labbe M, Michaud F., Online global loop closure detection for large-scale\r\nmulti-session graph-based slam, IEEE Intelligent Robots and Systems\r\n(IROS), 2014.\r\n[10] Labb M., Michaud F., Memory management for real-time\r\nappearance-based loop closure detection, IEEE Intelligent Robots\r\nand Systems (IROS), 2011.\r\n[11] Rublee, E., Rabaud, V., Konolige, K., et al., ORB: an efficient alternative\r\nto SIFT or SURF, IEEE International Conference on Computer Vision,\r\n2011.\r\n[12] Zhong, Y., Intrinsic shape signatures: A shape descriptor for 3D\r\nobject recognition, IEEE International Conference on Computer Vision\r\nWorkshops, 2009.\r\n[13] Glvez-Lpez, D., Tardos, J. D., Bags of binary words for fast place\r\nrecognition in image sequences(J). IEEE Transactions on Robotics, 2012,\r\n28(5): 1188-1197.","publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 123, 2017"}