Skip to main navigation Skip to main content

Korean. J. Breed. Sci. : Korean Journal of Breeding Science

OPEN ACCESS
ABOUT
BROWSE ARTICLES
EDITORIAL POLICIES
FOR CONTRIBUTORS

Articles

Article

YOLOv8 딥러닝 모델을 통한 새싹 채소의 부위별 길이 측정

조영훈1, 김재윤2, 하정민1,3,*

Application of Deep Learning Technology for Phenotyping Tissue Specific Length of Sprout Vegetables Using YOLOv8

Korean Journal of Breeding Science 2024;56(4):417-424.
Published online: December 1, 2024

1서울대학교 농림생물자원학부, 농업생명과학연구원 식물생명과학연구소

2국립공주대학교 식물자원학과

3서울대학교 식물유전체육종연구소

1Department of Agriculture, Forestry and Bioresources and Research Institute of Agriculture and Life Sciences, Seoul National University, Seoul 08826, Republic of Korea

2Department of Plant Resources, College of Industrial Science, Kongju National University, Chungcheongnam-do 32439, Republic of Korea

3Crop Genomics Lab, Plant Genomics and Breeding Institute, Seoul National University, Rm. 4105 Bldg. 200 CALS, 1 Gwanak‑ro, Gwanak‑gu, Seoul 08826, Republic of Korea

*Corresponding to Jungmin HaTEL. +82-2-880-4545E-mail. jungmin.ha@snu.ac.kr
• Received: September 6, 2024   • Revised: September 27, 2024   • Accepted: October 9, 2024

Copyright © 2024 by the Korean Society of Breeding Science

This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 132 Views
  • 1 Download
  • 1 Crossref
prev next

Citations

Citations to this article as recorded by  Crossref logo
  • Machine Learning Method to Select Single Nucleotide Polymorphism Markers for Protein Content, Grain Filling Rate, Height, and Panicle Length in Korean Rice
    Jeong-Gu Kim, Minwoo Kim, Gyu-Hwang Park, Jinhyun Kim, Jinho Jung, Tae-Ho Lee
    Korean Journal of Breeding Science.2025; 57(4): 403.     CrossRef

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

Include:

Application of Deep Learning Technology for Phenotyping Tissue Specific Length of Sprout Vegetables Using YOLOv8
Korean. J. Breed. Sci.. 2024;56(4):417-424.   Published online December 1, 2024
Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:
Include:
Application of Deep Learning Technology for Phenotyping Tissue Specific Length of Sprout Vegetables Using YOLOv8
Korean. J. Breed. Sci.. 2024;56(4):417-424.   Published online December 1, 2024
Close

Figure

  • 0
  • 1
  • 2
  • 3
  • 4
  • 5
Application of Deep Learning Technology for Phenotyping Tissue Specific Length of Sprout Vegetables Using YOLOv8
Image Image Image Image Image Image
Fig. 1 Image labeling for segmentation learning.
Fig. 2 Graphs of training metrics. The decline in loss values and the increase in precision, recall, and mAP values over epochs indicate that the training has been well conducted.
Fig. 3 Detected hypocotyl and root using deep learning model. Confidence scores for detection are indicated for each box.
Fig. 4 Length measuring function by deep leaning model. Approximation length is indicated for each box.
Fig. 5 Detected hypocotyl and root in various sprout vegetable (Williams 82, 독새기콩, 풍산나물콩, Vu365, Vu81).
Fig. 6 The optimal condition for the image detection and segmentation. (A) Ideal image. (B) Detection result of ideal image. (C) Non-Ideal image. (D) Detection result of Non-Ideal image.
Application of Deep Learning Technology for Phenotyping Tissue Specific Length of Sprout Vegetables Using YOLOv8

Number of images in the datasets produced using Roboflow.

Types of Dataset Set 1 Set 2
Raw Image 21 26
Augmented Image 30 30
Train 45 45
Valid 4 7
Test 2 4

Specification of images used in machine learning.

Item Specifications
Image size 640×640
Batch 16
Epochs 500
Patience 100

Approximate length of hypocotyl and root measured by deep learning model.

Number Hypocotyl (cm) Root (cm)
1 9.3 5.5
2 8.5 4.7
3 7.6 2.9
4 6.7 2
5 7.3 4.2
6 7.9 5.1
7 7.5 5.2
8 6.6 4
9 7.4 8.1
10 8.1 6.5
Table 1 Number of images in the datasets produced using Roboflow.
Table 2 Specification of images used in machine learning.
Table 3 Approximate length of hypocotyl and root measured by deep learning model.