Metrological Analysis Of Microsoft Kinect In The Context Of Object Localization

• Autorzy: Skalski A. , Machura B.

Identyfikatory

DOI

10.1515/mms-2015-0050

• Języki publikacji: EN

Abstrakty

EN

This paper presents a comprehensive metrological analysis of the Microsoft Kinect motion sensor performed using a proprietary flat marker. The designed marker was used to estimate its position in the external coordinate system associated with the sensor. The study includes calibration of the RGB and IR cameras, parameter identification and image registration. The metrological analysis is based on the data corrected for sensor optical distortions. From the metrological point of view, localization errors are related to the distance of an object from the sensor. Therefore, the rotation angles were determined and an accuracy assessment of the depth maps was performed. The analysis was carried out for the distances from the marker in the range of 0.8-1.65 m. The maximum average error was equal to 23 mm for the distance of 1.6 m.

Słowa kluczowe

EN

Kinect motion sensor; localization; camera calibration; measurement precision; robustness

• Wydawca: Komitet Metrologii i Aparatury Naukowej PAN
• Czasopismo: Metrology and Measurement Systems
• Rocznik: 2015
• Tom: Vol. 22, nr 4
• Strony: 469--478
• Opis fizyczny: Bibliogr. 17 poz., fot., rys., tab., wykr., wzory

Twórcy

autor

Skalski A.

• AGH University of Science and Technology, Department of Measurement and Electronics, Al. Mickiewicza 30, 30-059 Kraków, Poland

autor

Machura B.

• AGH University of Science and Technology, Department of Measurement and Electronics, Al. Mickiewicza 30, 30-059 Kraków, Poland

Bibliografia

• [1] Berger, K., et al. (2013). A state of the art report on kinect sensor setups in computer vision. Time-of-Flight and Depth Imaging. Sensors, Algorithms, and Applications, Springer Berlin Heidelberg, 257-272.
• [2] Pedro, L.M., de Paula Caurin, G.A. (2012). Kinect evaluation for human body movement analysis, Biomedical Robotics and Biomechatronics (BioRob). 4th IEEE RAS & EMBS International Conference on, 24-27, 1856-1861.
• [3] Kar, A. (2010). Skeletal tracking using microsoft kinect. Methodology, 1, 1-11.
• [4] Ross, A.C., et al. (2012). Validity of the Microsoft Kinect for assessment of postural control. Gait & posture, 36(3), 372-377.
• [5] Baek-Lok, O., et al. (2014). Validity and reliability of head posture measurement using Microsoft Kinect. British Journal of Ophthalmology, 98(11), 1560-1564.
• [6] Yu, C., Verkhoglyad, A., Poleshchuk, A., et al. (2014). 3D Optical Measuring Systems and Laser Technologies for Scientific and Industrial Applications. Measurement Science Review, 13(6), 322-328.
• [7] Toth, K.C., et al. (2012). Calibrating the MS Kinect Sensor. ASPRS Annual Conference, Sacramento, USA.
• [8] Khoshelham, K., Elberink, S.O. (2012). Accuracy and resolution of kinect depth data for indoor mapping applications. Sensors, 12(2), 1437-1454.
• [9] Gonzalez-Jorge, H., et al. (2013). Metrological evaluation of Microsoft Kinect and Asus Xtion sensors. Measurement, 46(6), 1800-1806.
• [10] Dutta, T. (2012). Evaluation of the Kinect™ sensor for 3-D kinematic measurement in the workplace. Applied Ergonomics, 43(4), 645-649.
• [11] Vicon system: http://www.vicon.com/
• [12] Fiala, M., Shu, Ch. (2008). Self-identifying patterns for plane-based camera calibration. Machine Vision and Applications, 19(4), 209-216.
• [13] Hartley, R., Zisserman, A. (2001). Multiple View Geometry in Computer Vision. Prentice Hall.
• [14] Stein, G.P. (1997). Lens distortion calibration using point correspondences, Computer Vision and Pattern Recognition. Proc. of 1997 IEEE Computer Society Conference on, 602-608.
• [15] Heikklä, J., Silven, O. (1997). A four-step camera calibration procedure with implicit image correction. Proc. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, San Juan, Puerto Rico, 1106-1112.
• [16] Andersen, M.R., et al. (2012). Kinect depth sensor evaluation for computer vision applications. Department of Engineering. Aarhus University. Denmark - Technical report ECE-TR-6.
• [17] Pascoal, P.B., et al. (2015). Retrieval of Objects Captured with Kinect One Camera. Eurographics Workshop on 3D Object Retrieval.

Uwagi

EN

The work was financed from the Dean Grants (statutory activity) and the AGH Rector Grant.