Implementation of 3D Location Detection Embedded System for Tomato Harvesting Robots 


Vol. 47,  No. 11, pp. 2007-2019, Nov.  2022
10.7840/kics.2022.47.11.2007


PDF
  Abstract

In recent years, the research on artificial intelligence-based agricultural automation technology has received a lot of attention to solve the problem of labor shortage due to the continuous decline of the agricultural population and aging. In this paper, we implement a deep learning-based three-dimensional location detection embedded system for the development of automated harvesting robots for tomatoes and aim to verify the applicability of the system in a real environment. The three-dimensional positioning system consists of NVIDIA Jetson Xavier NX, a low-power, small and low cost, and a stereo-type ZED2 camera to obtain three-dimensional information on images. The proposed system learns tomato images using the latest YOLOv5 object detection model and converts the two-dimensional coordinates of the detected tomatoes into three-dimensional coordinates through the learned model to detect the three-dimensional position of the tomatoes. To improve the performance of the trained model, we also apply TensorRT, a model optimization engine that can improve inference speed from several to dozens of times. The performance of the implemented system was compared with the average precision and image inference time of the optimization model applied with TensorRT, and the performance improvement was confirmed.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Related Articles
  Cite this article

[IEEE Style]

K. Lee, Y. Kim, B. Cho, W. Kim, M. Kim, Y. Hong, K. Kim, "Implementation of 3D Location Detection Embedded System for Tomato Harvesting Robots," The Journal of Korean Institute of Communications and Information Sciences, vol. 47, no. 11, pp. 2007-2019, 2022. DOI: 10.7840/kics.2022.47.11.2007.

[ACM Style]

Ki-Beom Lee, Yong-Hyun Kim, Byeong-Hyo Cho, Won-Kyung Kim, Man-Jung Kim, Youngki Hong, and Kyoung-Chul Kim. 2022. Implementation of 3D Location Detection Embedded System for Tomato Harvesting Robots. The Journal of Korean Institute of Communications and Information Sciences, 47, 11, (2022), 2007-2019. DOI: 10.7840/kics.2022.47.11.2007.

[KICS Style]

Ki-Beom Lee, Yong-Hyun Kim, Byeong-Hyo Cho, Won-Kyung Kim, Man-Jung Kim, Youngki Hong, Kyoung-Chul Kim, "Implementation of 3D Location Detection Embedded System for Tomato Harvesting Robots," The Journal of Korean Institute of Communications and Information Sciences, vol. 47, no. 11, pp. 2007-2019, 11. 2022. (https://doi.org/10.7840/kics.2022.47.11.2007)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880