DOI QR코드

DOI QR Code

Estimation of fruit number of apple tree based on YOLOv5 and regression model

YOLOv5 및 다항 회귀 모델을 활용한 사과나무의 착과량 예측 방법

  • Hee-Jin Gwak (Dept. of Computer Engineering, Andong National University) ;
  • Yunju Jeong (SW Expert Training Center, Andong National University) ;
  • Ik-Jo Chun (Dept. of Smart Horticultural Science, Andong National University) ;
  • Cheol-Hee Lee (Dept. of Computer Engineering, Andong National University)
  • 곽희진 ;
  • 정윤주 ;
  • 전익조 ;
  • 이철희
  • Received : 2024.05.10
  • Accepted : 2024.06.17
  • Published : 2024.06.30

Abstract

In this paper, we propose a novel algorithm for predicting the number of apples on an apple tree using a deep learning-based object detection model and a polynomial regression model. Measuring the number of apples on an apple tree can be used to predict apple yield and to assess losses for determining agricultural disaster insurance payouts. To measure apple fruit load, we photographed the front and back sides of apple trees. We manually labeled the apples in the captured images to construct a dataset, which was then used to train a one-stage object detection CNN model. However, when apples on an apple tree are obscured by leaves, branches, or other parts of the tree, they may not be captured in images. Consequently, it becomes difficult for image recognition-based deep learning models to detect or infer the presence of these apples. To address this issue, we propose a two-stage inference process. In the first stage, we utilize an image-based deep learning model to count the number of apples in photos taken from both sides of the apple tree. In the second stage, we conduct a polynomial regression analysis, using the total apple count from the deep learning model as the independent variable, and the actual number of apples manually counted during an on-site visit to the orchard as the dependent variable. The performance evaluation of the two-stage inference system proposed in this paper showed an average accuracy of 90.98% in counting the number of apples on each apple tree. Therefore, the proposed method can significantly reduce the time and cost associated with manually counting apples. Furthermore, this approach has the potential to be widely adopted as a new foundational technology for fruit load estimation in related fields using deep learning.

본 논문은 딥러닝 기반 객체 탐지 모델과 다항 회귀모델을 이용하여 사과나무에 열린 사과의 개수를 예측할 수 있는 새로운 알고리즘을 제안한다. 사과나무에 열린 사과의 개수를 측정하면 사과 생산량을 예측할 수 있고, 농산물 재해 보험금 산정을 위한 손실을 평가하는 데에도 활용할 수 있다. 사과 착과량 측정을 위해 사과나무의 앞면과 뒷면을 촬영하였다. 촬영된 사진에서 사과를 식별하여 라벨링한 데이터 세트를 구축하였고, 이 데이터 세트를 활용하여 1단계 객체 탐지 방식의 CNN 모델을 학습시켰다. 그런데 사과나무에서 사과가 나뭇잎, 가지 등으로 가려진 경우 영상에 포착되지 않아 영상 인식 기반의 딥러닝 모델이 해당 사과를 인식하거나 추론하는 것이 어렵다. 이 문제를 해결하기 위해, 우리는 두 단계로 이루어진 추론 과정을 제안한다. 첫 번째 단계에서는 영상 기반 딥러닝 모델을 사용하여 사과나무의 양쪽에서 촬영한 사진에서 각각의 사과 개수를 측정한다. 두 번째 단계에서는 딥러닝 모델로 측정한 사과 개수의 합을 독립변수로, 사람이 실제로 과수원을 방문하여 카운트한 사과 개수를 종속변수로 설정하여 다항 회귀 분석을 수행한다. 본 논문에서 제안하는 2단계 추론 시스템의 성능 평가 결과, 각 사과나무에서 사과 개수를 측정하는 평균 정확도가 90.98%로 나타났다. 따라서 제안된 방법은 수작업으로 사과의 개수를 측정하는 데 드는 시간과 비용을 크게 절감할 수 있다. 또한, 이 방법은 딥러닝 기반 착과량 예측의 새로운 기반 기술로 관련 분야에서 널리 활용될 수 있을 것이다.

Keywords

Acknowledgement

This work was supported by a Research Grant of Andong National University

References

  1. Yildirim, Sahin and Burak Ulu, "Deep learning based apples counting for yield forecast using proposed flying robotic system," Sensors 2023, 23(13), 6171, 2023. DOI: https://doi.org/10.3390/s23136171 
  2. Ross Girshick, Jeff Donahue, Trevor Darrell and Jitendra Malik, "Rich feature hierarchies for accurate object detection and semantic segmentation," In Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 580-587, 2014. DOI: https://doi.org/10.48550/arXiv.1311.2524 
  3. Ross Girshick, "Fast R-CNN," In Proceedings of the IEEE international conference on computer vision, pp.1440-1448, 2015. DOI: https://doi.org/10.48550/arXiv.1504.08083 
  4. Shaoqing Ren, Kaiming He, Ross Girshick and Jian Sun, "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks," Advances in neural information processing systems, 28, 2015. DOI: https://doi.org/10.48550/arXiv.1506.01497 
  5. Kaiming He, Georgia Gkioxari, Piotr Dollar and Ross Girshick, "Mask R-CNN," In Proceedings of the IEEE international conference on computer vision, pp.2961-2969, 2017. DOI: https://doi.org/10.48550/arXiv.1703.06870 
  6. Joseph Redmon, Santosh Divvala, Ross Girshick and Ali Farhadi, "You Only Look Once: Unified, Real-Time Object Detection," In Proceedings of the IEEE conference on computer vision and pattern recognition, pp.779-788, 2016. DOI: https://doi.org/10.48550/arXiv.1506.02640 
  7. LIU, Wei, et al. "Ssd: Single shot multibox detector," In Computer Vision-ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11-14, 2016, Proceedings, Part I 14, Springer International Publishing, p. 21-37, 2016. DOI: https://doi.org/10.48550/arXiv.1512.02325 
  8. Bochkovskiy, Alexey, Chien-Yao Wang, and Hong-Yuan Mark Liao. "Yolov4: Optimal speed and accuracy of object detection," arXiv preprint arXiv:2004.10934 (2020). DOI: https://doi.org/10.48550/arXiv.2004.10934 
  9. "Comprehensive Guide to Ultralytics YOLOv5," https://docs.ultralytics.com/yolov5/ 
  10. Wang, Chien-Yao, Alexey Bochkovskiy, and Hong-Yuan Mark Liao, "YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors." Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.7464-7475, 2023. DOI: https://doi.org/10.48550/arXiv.2207.02696