Journal of Drive and Control (드라이브 ㆍ 컨트롤)

The Korean Society for Fluid Power and Construction Equipment (유공압건설기계학회)
권호 표지
DOI QR코드

DOI QR Code

Estimation of two-dimensional position of soybean crop for developing weeding robot

제초로봇 개발을 위한 2차원 콩 작물 위치 자동검출

  • SooHyun Cho (Department of Biosystem Machinery Engineering, Chungnam National University) ;
  • ChungYeol Lee (Koreatriaxle Co. Ltd) ;
  • HeeJong Jeong (Koreatriaxle Co. Ltd) ;
  • SeungWoo Kang (Department of Biosystem Machinery Engineering, Chungnam National University) ;
  • DaeHyun Lee (Department of Biosystem Machinery Engineering, Chungnam National University)
  • Received : 2023.03.09
  • Accepted : 2023.05.04
  • Published : 2023.06.01
Download PDF
⟨ Previous Next ⟩

Abstract

In this study, two-dimensional location of crops for auto weeding was detected using deep learning. To construct a dataset for soybean detection, an image-capturing system was developed using a mono camera and single-board computer and the system was mounted on a weeding robot to collect soybean images. A dataset was constructed by extracting RoI (region of interest) from the raw image and each sample was labeled with soybean and the background for classification learning. The deep learning model consisted of four convolutional layers and was trained with a weakly supervised learning method that can provide object localization only using image-level labeling. Localization of the soybean area can be visualized via CAM and the two-dimensional position of the soybean was estimated by clustering the pixels associated with the soybean area and transforming the pixel coordinates to world coordinates. The actual position, which is determined manually as pixel coordinates in the image was evaluated and performances were 6.6(X-axis), 5.1(Y-axis) and 1.2(X-axis), 2.2(Y-axis) for MSE and RMSE about world coordinates, respectively. From the results, we confirmed that the center position of the soybean area derived through deep learning was sufficient for use in automatic weeding systems.

Keywords

Acknowledgement

본 과제(결과물)는 농림축산식품부의 재원으로 농림식품기술기획평가원의 첨단농기계산업화기술개발사업(321061-2)의 지원을 받아 연구되었으며, 2023년도 교육부의 재원으로 한국연구재단의 지원을 받아 수행된 지자체-대학 협력기반 지역혁신 사업의 결과임(2021RIS-004).

References

  1. S. Y. Baek, et. al., "Design Verification of an E-driving System of a 44 kW-class Electric Tractor using Agricultural Workload Data", Journal of Drive and Control, Vol.19, No.4 pp36-45, 2022
  2. T. Bakker, "An autonomous robot for weed control: Design, navigation and control," Wageningen, Netherland, 2009.
  3. O. Bawden, et. al., "Robot for weed species plant-specific management.", Journal of Field Robotics, Vol.34, No.6, pp.1179-1199, 2017. https://doi.org/10.1002/rob.21727
  4. K. N. Harker, J.T. O'Donovan, "Recent weed control, weed management, and integrated weed management.", Weed Technology, Vol.27, No.1, pp.1-11, 2013. https://doi.org/10.1614/WT-D-12-00109.1
  5. JBARES, 2018, Soybean cultivation technology, Jeollabuk-do Agricultural Technology and Extension Services, Iksan, 108 pp.
  6. D. K. Noh, et. al., "Analysis of Surplus Flow in a Hydraulic System Applied to a Self-propelled Spinach Harvester", Journal of Drive and Control, Vol.19, No.1 pp26-33, 2022. https://doi.org/10.7839/KSFC.2022.19.1.026
  7. W. S. Kim, et. al., "Stereo-vision-based crop height estimation for agricultural robots.", Computers and Electronics in Agriculture, 181, 2021.
  8. KOSIS, Production cost survey of agricultural and livestock products, 2021, 2023.03.08, Labor input time by job, 'Labor input time by province/work', Korean Statistical Information Service.
  9. Y. T. Kim, et. al., "Technology trend on autonomous agricultural machinery.", J. of Drive and Control, Vol.19, No.1, pp.95-99, 2022.
  10. D. H. Lee et al, "Study on image-based flock density evaluation of broiler chicks", Journal of Korea Institute of Information, Electronics, and Communication Technology, Vol.12, Issue.4, pp.373-379, 2019.
  11. J. I. Lee et. al., "Development of Virtual Simulator and Database for Deep Learning-based Object Detection", Journal of Drive and Control, Vol.18 No.4 pp.9-18, 2021. https://doi.org/10.7839/KSFC.2021.18.4.009
  12. Y. Li, et. al., "Key technologies of machine vision for weeding robots: A review and benchmark.", Computers and Electronics in Agriculture, 2022.
  13. R. Raffik, et. al., "Autonomous Weeding Robot for Organic Farming Fields.", International Conference on Advancements in Electrical, Electronics, Communication, Computing and Automation, pp.1-4, 2021.
  14. RDA, 2018, "Introduction of the occurrence status and control method of 'malicious weed'.", Rural Development Administration, Jeonju.
  15. F. Sellmann, et. al., "RemoteFarming 1: Human-machine interaction for a field-robot-based weed control application in organic farming.", In 4th International Conference on Machine Control & Guidance, pp.19-20, 2014.
  16. S. Y. Shin, et. al., "Basic study on the development of weeding robot for upland field (1)-Analysis of weeding effect for a yield of vegetable.", Proceeding of the KSAM & ARCs 2021 Autumn Conference, Vol.26, No.2, pp.218-218, 2021.
  17. Wolfe, J. William, et. al., "The perspective view of three Points", IEEE Transactions on Pattern Analysis and machine intelligence, Vol.12, No.1, pp.66-73, 1991. https://doi.org/10.1109/34.67632
  18. J. H. Won, et. al., "Study on Traveling Characteristics of Straight Automatic Steering Devices for Drivable Agricultural Machinery", Journal of Drive and Control, Vol.19, No.4 pp19-28, 2022. https://doi.org/10.7839/KSFC.2022.19.4.019
  19. Z. Zhang, "A flexible new technique for camera calibration.", IEEE Transactions on pattern analysis and machine intelligence, Vol.22, No.11, pp.1330-1334, 2000. https://doi.org/10.1109/34.888718
  20. B. Zhou, et. al., "Learning deep features for discriminative localization", In Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 2921-2929, 2016.