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

디지털육종을 위한 RGB 이미지 기반 사과 과실 형태 측정 최적화 연구

유재일1, 이채원2, 이서연1, 김영욱1, 김년희1, 송지선1, 백정호1, 조정건3, 김경환1,*

Optimization Study of RGB Image-based Apple Fruit Measurement for Digital Breeding

Korean Journal of Breeding Science 2023;55(4):303-310.
Published online: December 1, 2023

1국립농업과학원 유전자공학과

2국립식량과학원 재배환경과

3국립원예특작과학원 과수과

1Gene Engineering Division, National Institute of Agricultural Sciences, Jeonju 54874, Republic of Korea

2Crop Cultivation & Environment Research Division, National Institute of Crop Sciences, Suwon 16613, Republic of Korea

3Fruit Research Division, National Institute of Horticultural and Herbal Science, Wanju 55365, Republic of Korea

*Corresponding Author (E-mail: biopiakim@korea.kr, Tel: +82-63-238-4651, Fax: +82-63-238-4654)
• Received: November 6, 2023   • Revised: November 9, 2023   • Accepted: November 15, 2023

Copyright © 2023 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.

  • 119 Views
  • 1 Download
  • 3 Crossref
prev next

Citations

Citations to this article as recorded by  Crossref logo
  • Soft-hard multiphase materials co-milling: Synergistic fragmentation, activation and mesoscale mixing of heterogeneous solid wastes
    Kai Lyu, Xinyu Chen, Jingkai Zhou, Lin Liu, Xiaoyan Liu, Jinna Shi
    Cement and Concrete Composites.2026; 173: 106747.     CrossRef
  • RETRACTED: Application of RSM- CCD methodology and image J. for modeling and optimization of orchid protocorm encapsulation
    Zahra Mahdavi, Shirin Dianati Daylami, Ali Fadavi, Mandana Mahfeli
    Heliyon.2025; 11(4): e42744.     CrossRef
  • Deep Learning-based Weight and Grade Prediction Algorithm from Apple Images
    Ju-Hwan Lee, Nguyen Bui Ngoc Han, Jin Lee, Le Bao Thai An, Gyeong-Ju Kwon, Jin-Young Kim
    The Journal of Korean Institute of Information Technology.2024; 22(6): 31.     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:

Optimization Study of RGB Image-based Apple Fruit Measurement for Digital Breeding
Korean. J. Breed. Sci.. 2023;55(4):303-310.   Published online December 1, 2023
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:
Optimization Study of RGB Image-based Apple Fruit Measurement for Digital Breeding
Korean. J. Breed. Sci.. 2023;55(4):303-310.   Published online December 1, 2023
Close

Figure

  • 0
  • 1
  • 2
  • 3
Optimization Study of RGB Image-based Apple Fruit Measurement for Digital Breeding
Image Image Image Image
Fig. 1 Photographs of the four apple cultivars used for image analysis in this study.
Fig. 2 Optimization of apple fruit image acquisition conditions based on scale bar position, lighting conditions, and background color variation. (A) Images from testing the three image acquisition conditions for optimization. (B) Validation of image acquisition conditions using scatter plot and masked images.
Fig. 3 Comparison of differences among individual fruits based on fruit roundness index values. (A) Comparing the maximum and minimum values of fruits using the roundness index. (B) PCA plot based on four index values including aspect ratio, roundness, circularity, solidity.
Fig. 4 Accuracy comparison of Top and Side image indices with actual length and width measurements of fruits. In this analysis, Vernier calipers were used for actual measurements, and a total of 30 fruits per cultivar were analyzed. The results were compared with six diameter-related indices.
Optimization Study of RGB Image-based Apple Fruit Measurement for Digital Breeding

Definitions of the 12 imageJ-based indices used in this study.

Category Index Description
Size Area Area of selection in square pixels or in calibrated square units
Perimeter The length of the outside boundary of the selection
Diameter Height Measure the height of a digitized object with ROI
Width Measure the width of a digitized object with ROI
Major Major axis (transverse diameter)
Minor Minor axis (conjugate diameter)
Feret The longest distance between any two points along the selection boundary, also known as maximum caliper.
MinFeret The shortest distance between any two points along the selection boundary, also known as minimum caliper.
Shape Aspect ratio The aspect ratio of the particle’s fitted ellipse, i.e., [Major Axis]/[Minor Axis]
Roundness 4×[Area]/(π×[Major axis]2 or the inverse of aspect ratio.
Circularity 4π×[Area]/[Perimeter]2 with a value of 1.0 indicating a perfect circle.
Solidity

[Area]/[Convex area]

*Convex hull can be thought of as a rubber band wrapped tightly around the points that define the selection, and determined by gift wrap algorithm

Variations in irradiance and illuminance based on four lighting conditions.

Lighting conditions Irradiance(W/m2) Illuminance(lux)
1 (light box) 3.32±0.20 1046.87±59.57
2 (light box+CN-T96) 11.66±0.36 3239.40±103.09
3 (light box+VL-D85T) 18.60±0.53 4965.40±142.66
4 (light box+CN-T96+VL-D85T) 26.56±0.47 7052.87±127.33

Mean values of 12 indices extracted from Top and Side fruit images for 4 Apple cultivars.

Index Image type Hongan Hongro Fuji Hwangok
Area Top 6160.68±597.04a 5343.83±665.38b 5468.55±927.4b 4143.31±385.01c
Side 3955.82±404.98a 3694.56±484.16b 3554.53±525.69b 2916.51±260.23c
Perimeter Top 297.73±14.98a 276.6±17.53b 278.57±23.99b 247.65±11.5c
Side 254.46±13.13a 240.92±14.78b 241.62±18.38b 221.53±10.46c
Width Top 87.88±4.49a 82.45±5.72b 82.89±7.39b 72.29±3.77c
Side 71.45±3.1a 67.61±4.18b 67.85±5.58b 60.23±2.87c
Height Top 89.54±4.34a 83.73±5.19b 84.23±7.26b 73.67±3.47c
Side 66.5±4.53a 66.15±4.66a 63.29±4.48b 58.83±3.36c
Major Top 90.14±4.33a 84.19±5.36b 85.43±7.5b 74.36±3.65c
Side 73.44±3.13a 70.12±4.65b 69.53±5.31b 62.58±2.63c
Minor Top 86.83±4.44a 80.53±4.95b 80.95±6.85b 70.8±3.39c
Side 68.43±4.49a 66.83±4.28ab 64.77±4.77b 59.23±3.05c
Feret Top 90.65±4.22a 85.18±5.6b 85.83±7.52b 74.88±3.6c
Side 75.23±3.87a 72.58±4.91b 71.09±5.13b 64.51±2.98c
MinFeret Top 86.65±4.54a 80.16±5.04b 80.86±6.82b 70.74±3.26c
Side 66.3±4.39a 65.47±4.14a 63.09±4.59b 58.21±3.25c
Circularity Top 0.8713±0.01a 0.8753±0.01ab 0.8797±0.01b 0.8477±0.02c
Side 0.77±0.03b 0.8±0.03a 0.76±0.03bc 0.75±0.04c
Aspect ratio Top 1.0393±0.02a 1.0463±0.02ab 1.0555±0.03b 1.0500±0.03ab
Side 1.0750±0.05a 1.0507±0.02b 1.0738±0.03a 1.0573±0.03ab
Roundness Top 0.9633±0.02a 0.9573±0.02ab 0.9483±0.03b 0.9530±0.02ab
Side 0.9320±0.04a 0.9533±0.02b 0.9321±0.03a 0.9467±0.03ab
Solidity Top 1.00±0.00 1.00±0.00 1.00±0.00 1.00±0.00
Side 0.99±0.00 0.99±0.00 0.99±0.00 0.99±0.00
Table 1 Definitions of the 12 imageJ-based indices used in this study.
Table 2 Variations in irradiance and illuminance based on four lighting conditions.
Table 3 Mean values of 12 indices extracted from Top and Side fruit images for 4 Apple cultivars.

*Values with different letters (a-c) are statistically different at p<0.05 significant level (post hoc Duncan’s test); data represent the means±SD (n=30).