Skip to main content Skip to main navigation menu Skip to site footer
Scientific Electronic Archives
Mathematical Sciences / Engineering
DOI: 10.36560/141120211467
Submitted: 2021-08-23
Published: 2021-10-29

Mathematical models for metric features extraction from RGB-D sensor

Universidade Federal de Mato Grosso
Embrapa Informática Agropecuária
Embrapa Agrossilvipastoril
Universidade Federal de Mato Grosso - Campus Sinop
University of Illinois
image processing; depth camera; RealSense™.

Abstract

The use of the RGB-D camera has been applied in several fields of science. That popularization is due to the emergence of technologies such as the Intel® RealSense™ D400 series. However, despite the actual demand from some potential users, few studies concern the characterization of these sensors for object measurements. Our study sought to estimate models dealing with calculating the area and length between targets or points within RGB and depth images.  An experiment was set up with white cardboard fixed on a flat surface with colored pins. We measured the distance between the camera and cardboard by calculating the average distance from the pixels belonging to the target area. The Information Criterion AIC and BIC associated with R2 were performed to select the best models. Polynomial and power regression models reached the highest coefficient of determination and smallest values of AIC and BIC.

References

  1. BANGLI, L.; HAIBIN, C.; ZHAOJIE, J.; HONGHAI, L. Bangli. L.; Haibin. C.; Zhaojie. J.; Honghai. L. RGB-D sensing-based human action and interaction analysis: a survey. Pattern Recognition, vol., 94, p. 1–12, 2019.
  2. BASSO, F. MUNARO, M.; MICHIELETTO, S. Basso, F.; Munaro, M.; Michieletto, S.; et al. Fast and robust multi-people tracking from RGB-D data for a mobile robot. Advances in Intelligent Systems and Computing, vol. 193, p. 265–276, 2013.
  3. CARFAGNI, M.; FURFERI, R.; GOVERNI, L.; SERVI, M.; UCCHEDDUU, F.; VOLPE, Y. On the Performance of the Intel SR300 Depth Camera: Metrological and Critical Characterization. In: IEEE Sensors Journal, vol. 17, p. 4508-4519, 2017. doi: 10.1109/JSEN.2017.2703829.
  4. CHOO, B.; DeVORE, M.D.; BELING, P.A. Statistical models of horizontal and vertical stochastic noise for the Microsoft Kinect™ sensor. In: IECON 2014 - 40th Annual Conference of the IEEE Industrial Electronics Society, Dallas, TX, 2014, p. 2624-2630, doi: 10.1109/IECON.2014.7048876.
  5. CONDOTTA, I.C.F.S.; BROWN-BRANDL, T.M.; SILVA-MIRANDA, K.O. Evaluation of using a depth sensor to estimate the weight of finishing pigs. In: 8th European Conference on Precision Livestock Farming, 2017, Nantes, 2017. vol. 1. p. 495-502.
  6. CONDOTTA, I.C.F.S.; BROWN-BRANDL, T.M.; SOUZA, R.V.; SILVA-MIRANDA, K.O. Using an artificial neural network to predict pig mass from depth images. In: 10th International Livestock Environment Symposium (ILES X), 2018, Omaha, NE. 2018.
  7. CONDOTTA, I.C.F.S.; BROWN-BRANDL, T.M.; SOUZA, R.V.; SILVA-MIRANDA, K.O.; STINN, J.P. Evaluation of a depth sensor for mass estimation of growing and finishing pigs. Biosyst. Eng., vol. 173, p. 11–18, 2018a. doi: 10.1016/j.biosystemseng.2018.03.002.
  8. CONDOTTA, I.C.F.S.; BROWN-BRANDL, T.M.; PITLA, S.K.; STINN, J.P.; SILVA-MIRANDA, K.O. Evaluation of low-cost depth cameras for agricultural applications. Computers and Electronics in Agriculture, vol.173, 2020
  9. Intel® Realsense™. User guide Intel Realsense D400 series/SR300 viewer. Revision 002, (2018).
  10. Intel® RealSense™. Lib. Realsense: D400 Series visual presets. 2.0 [San Francisco]: GitHub, 2019. 1 Phone code library. Available in: < https://github.com/IntelRealSense/librealsense/tree/master/examples/align < Access in: 02 mar 2020.
  11. KHOSHELHAM, K.; ELBERINK, S.O. Accuracy and resolution of Kinect depth data for indoor mapping applications. Sensors, vol. 12, p. 1437-1454, 2012.
  12. KONGSO, J. Estimation of pig weight using a Microsoft Kinect prototype imaging system. Computers and Electronics in Agriculture, vol. 39, p. 32–35, 2014.
  13. LAI, K.; BO, L.; REN, X.; FOX, D. A large-scale hierarchical multi-view RGB-D object dataset. In: IEEE International Conference on Robotics and Automation, Shanghai, 2011, p. 1817-1824. doi: 10.1109/ICRA.2011.5980382.
  14. NGUYEN, T.V.; FENG, J.; YAN, S. Seeing human weight from a single RGB-D image. Journal of Computer Science and Technology, vol. 29, p. 777–784, 2014.
  15. PEZZUOLO, A.; GUARINO, M.; SARTORI, L.; GONZÁLEZ, L.A. On-barn pig weight estimation based on body measurements by a Kinect v1 depth camera. Computers and Electronics in Agriculture, vol. 148, p. 29-36, 2018.
  16. VIT, A.; SHANI, G. Comparing RGB-D Sensors for Close Range Outdoor Agricultural Phenotyping. Sensors. 2018, 18, 4413.
  17. YU, H.; LEE, K.; MOROTA, G. Forecasting dynamic body weight of non-restrained pigs from images using an RGB-D sensor camera. Translational Animal Science. 2021, vol.5:1, txab006, https://doi.org/10.1093/tas/txab006
  18. WANG, S.; PAN, H.; ZHANG, C.; TIAN, Y. RGB-D image-based detection of stairs, pedestrian crosswalks and traffic signs. J. Vis. Commun. Image Representation, vol. 25, p. 263-272, 2014.

How to Cite

dos Santos, E. F. ., Vendrusculo, L. G. ., Lopes, L. B. ., Kamchen, S. G. ., & Condotta, I. C. F. S. . (2021). Mathematical models for metric features extraction from RGB-D sensor . Scientific Electronic Archives, 14(11). https://doi.org/10.36560/141120211467