Depth map characterization of RGB-D sensor for obstacle detection system

This paper presents a characterization of the depth map from an RGB-D sensor, Kinect 360, to be used as an input to an obstacle detection system. Since the accuracy, reliability and timeliness of the data are crucial in the system, there is a need to characterize it. Sensor repeatability tests are done to characterize the depth map. The mean and standard deviation of measured depths and mean error were computed to describe its detection accuracy and precision. Evident noises during tests were defined.

[1]  Ling Shao,et al.  Enhanced Computer Vision With Microsoft Kinect Sensor: A Review , 2013, IEEE Transactions on Cybernetics.

[2]  Berthold K. P. Horn.  Robot vision , 1986, MIT electrical engineering and computer science series.

[3]  Liang-Gee Chen,et al.  An intelligent depth-based obstacle detection system for visually-impaired aid applications , 2012, 2012 13th International Workshop on Image Analysis for Multimedia Interactive Services.

[4]  Charles K. Toth,et al.  ACCURACY TEST OF MICROSOFT KINECT FOR HUMAN MORPHOLOGIC MEASUREMENTS , 2012 .

[5]  Jidong Huang,et al.  Study on the use of Microsoft Kinect for robotics applications , 2012, Proceedings of the 2012 IEEE/ION Position, Location and Navigation Symposium.

[6]  Ka Wah Joe Lo.  Mobile robot = 機械人 , 2016 .

[7]  Y. J. Tejwani,et al.  Robot vision , 1989, IEEE International Symposium on Circuits and Systems,.

[8]  Partha Pratim Das,et al.  Characterizations of Noise in Kinect Depth Images: A Review , 2014, IEEE Sensors Journal.