Online First

Research Article

Non-destructive Analysis of Anthocyanin Content Levels Based on RGB Images of Red Lettuce

RGB 이미지 기반 적상추 안토시아닌 비파괴적 분석

Giju Eoh1
,
Sungsu Lim12
,
Jongseok Park13*
어 기주1
,
임 성수12
,
박 종석13*
1Graduate school of Bio-AI Convergence, Chungnam National University, Daejeon 34134, Korea
2Department of Computer Science and Engineering, Chungnam National University, Daejeon 34134, Korea
3Department of Horticultural Science, Chungnam National University, Daejeon 34134, Korea
1충남대학교 바이오AI융합학과
2충남대학교 컴퓨터융합학과
3충남대학교 원예학과
*Corresponding Author
PDF XML Citation
Abstract

The purpose of this study is to develop an optimal prediction model of the anthocyanin content of red lettuce (Lactuca sativa L. ‘Seonpungpochap’) with RGB image data for a non-destructive analysis method. An experiment was conducted to obtain a range of data on phenotypic red lettuce. LED light sources were used with the lettuce at ratios of 100:0 (R:B), 70:30, 50:50, 30:70, and 0:100, which represent the five treatments conducted here under a light intensity level of 180 ± 10 µmol·m-2·s-1. Images of red lettuce samples were obtained with an RGB camera, and the total anthocyanin content of each of the red lettuce samples for were analyzed using HPLC (high-performance liquid chromatography). The background of each of the lettuce images was removed with the ‘GrabCut’ algorithm, and the silhouette index was employed to identify the optimal K-value. A color histogram was constructed using the ‘K-means clustering’ algorithm. Multiple regression was conducted using the ‘B’, ‘RB/(R+G+B)’ ‘Brightest, 2G-R’ variables selected from the color histogram due to their strong correlation with the dependent variable. The result was a model with a high coefficient of determination of 0.953 (RMSE: 0.209). Hence, this study offers a means by which to conduct a non-destructive analysis of the anthocyanin content in red lettuce using an AI model, which can be utilized in plant factories and farms as a new indicator of red lettuce cultivation.

Keywords
computer vision
K-means clustering
Lactuca sativa
machine learning
plant factory

본 연구의 목적은 RGB 이미지 기반의 비파괴적인 방법으로 적상추(Lactuca sativa L. ‘선풍포찹’)의 안토시아닌 함량 예측 모델을 개발하는 것이다. 다양한 표현형의 적상추 데이터를 얻기 위해 적색과 청색으로 구성된 LED 광원을 사용하여 광도 180 ± 10µmol·m-2·s-1에서 비율 100:0(적:청), 70:30, 50:50, 30:70, 0:100의 5개 처리조건에서 재배했다. 수확 후, RGB 카메라를 이용하여 이미지를 얻고 액체 크로마토그래피(High-performance liquid chromatography, HPLC)를 이용하여 각 개체별 총 안토시아닌 함량 데이터를 분석했다. 상추 이미지의 배경은 ‘GrabCut’ 알고리즘을 사용하여 제거하고, Silhouette index를 사용하여 최적의 K값을 얻어 컬러 히스토그램을 구성했다. 색상 히스토그램에서 얻을 수 있는 상관 관계가 높은 변수 중 가장 높은 결정계수를 가지는 조합인 ‘B’, ‘RB/(R+G+B)’, ‘Brightest, 2G-R’ 3가지 변수를 독립변수로 사용하여 다중 회귀 분석한 결과 0.953(RMSE: 0.209) 라는 높은 결정계수를 가진 모델을 얻을 수 있었다. 본 연구는 인공지능을 활용하여 식물공장에서 이용할 수 있는 적상추 안토시아닌 함량 비파괴적 분석을 위한 모델을 제시했으며, 이는 식물공장과 농장에서 적상추 재배의 새로운 지표로 활용될 것으로 판단된다.

키워드
Lactuca sativa
식물공장
K-means clustering
컴퓨터비전
기계학습
References
1

Aranganayagi S, Thangavel K (2007) Clustering categorical data using silhouette coefficient as a relocating measure. Paper presented at the International conference on computational intelligence and multimedia applications (ICCIMA 2007). https://doi.org/10.1109/ICCIMA.2007.328

10.1109/ICCIMA.2007.328
2

Baek Y, Sul S, Cho YY (2023) Estimation of days after transplanting using an artificial intelligence CNN (convolutional neural network) model in a closed-type plant factory. Hortic Sci Technol 41:81-90. https://doi.org/10.7235/HORT.20230008

10.7235/HORT.20230008
3

Boldt JK, Meyer MH, Erwin JE (2014) Foliar anthocyanins: A horticultural review. Hortic Rev 42:209-252. https://doi.org/10.1002/9781118916827.ch04

10.1002/9781118916827.ch04
4

Boykov YY, Jolly MP (2001) Interactive graph cuts for optimal boundary and region segmentation of objects in ND images. In: Proceedings of the 8th IEEE International Conference on Computer Vision (ICCV), vol. 1, pp. 105-112. https://doi.org/10.1109/ICCV.2001.937505

10.1109/ICCV.2001.937505
5

Buxbaum N, Lieth JH, Earles JM (2022) Non-destructive plant biomass monitoring with high spatio-temporal resolution via proximal RGB-d imagery and end-to-end deep learning. Front Plant Sci 13:758818. https://doi.org/10.3389/fpls.2022.758818

10.3389/fpls.2022.75881835498682PMC9043900
6

Cavallo DP, Cefola M, Pace B, Logrieco AF, Attolico G (2017) Contactless and non-destructive chlorophyll content prediction by random forest regression: A case study on fresh-cut rocket leaves. Comput Electron Agric 140:303-310. https://doi.org/10.1016/j.compag.2017.06.012

10.1016/j.compag.2017.06.012
7

Choi SY, Lee AK (2020) Development of a cut rose longevity prediction model using thermography and machine learning. Hortic Sci Technol 38:675-685. https://doi.org/10.7235/HORT.20200061

10.7235/HORT.20200061
8

Chowdhury M, Ngo VD, Islam MN, Ali M, Islam S, Rasool K, Park SU, Chung SO (2021) Estimation of glucosinolates and anthocyanins in kale leaves grown in a plant factory using spectral reflectance. Horticulturae 7:56. https://doi.org/10.3390/horticulturae7030056

10.3390/horticulturae7030056
9

Del Valle JC, Gallardo-López A, Buide ML, Whittall JB, Narbona E (2018) Digital photography provides a fast, reliable, and noninvasive method to estimate anthocyanin pigment concentration in reproductive and vegetative plant tissues. Ecol Evol 8:3064-3076. https://doi.org/10.1002/ece3.3804

10.1002/ece3.380429607006PMC5869271
10

Gehan MA, Fahlgren N, Abbasi A, Berry J, Callen ST, Chavez L, Doust AN, Feldman M, Gilbert KB, et al. (2017) PlantCV v2: Image analysis software for high-throughput plant phenotyping. Peer J 5. https://doi.org/10.7717/peerj.4088

10.7717/peerj.408829209576PMC5713628
11

He R, Zhang Y, Song S, Su W, Hao Y, Liu H (2021) UV-A and FR irradiation improves growth and nutritional properties of lettuce grown in an artificial light plant factory. Food Chem 345:128727. https://doi.org/10.1016/j.foodchem.2020.128727

10.1016/j.foodchem.2020.12872733307433
12

Heo JW, Baek JH (2021) Effects of light-quality control on the plant growth in a plant factory system of artificial light type. Korean J Environ Agric 40:270-278. https://doi.org/10.5338/KJEA.2021.40.4.31

10.5338/KJEA.2021.40.4.31
13

Hwang IC, Noh HS, Yang DI, Kim MB (2021) Prediction of Paprika yield using multiple linear regression. JKICS 46:2048-2055. https://doi.org/10.7840/kics.2021.46.11.2048

10.7840/kics.2021.46.11.2048
14

Kanchanomai C, Ohashi S, Naphrom D, Nemoto W, Maniwara P, Nakano K (2020) Non-destructive analysis of Japanese table grape qualities using near-infrared spectroscopy. Hortic Environ Biotechnol 61:725-733. https://doi.org/10.1007/s13580-020-00256-4

10.1007/s13580-020-00256-4
15

Karki S, Basak JK, Paudel B, Deb NC, Kim NE, Kook J, Kang MY, Kim HT (2024) Classification of strawberry ripeness stages using machine learning algorithms and colour spaces. Hortic Environ Biotechnol 65:337-354. https://doi.org/10.1007/s13580-023-00559-2

10.1007/s13580-023-00559-2
16

Mochida K, Koda S, Inoue K, Hirayama T, Tanaka Y, Nishii T, Melgani F (2019) Computer vision-based phenotyping for improvement of plant productivity: A machine learning perspective. Gigascience 8. https://doi.org/10.1093/gigascience/giy153

10.1093/gigascience/giy15330520975PMC6312910
17

Mogol BA, Gökmen V (2014) Computer vision-based analysis of foods: A non-destructive colour measurement tool to monitor quality and safety. J Sci Food Agric 94:1259-1263. https://doi.org/10.1002/jsfa.6500

10.1002/jsfa.650024288215
18

Mustaffa IB, Khairul SF (2017) Identification of fruit size and maturity through fruit images using OpenCV-Python and Rasberry Pi. Paper presented at the 2017 International Conference on Robotics, Automation and Sciences (ICORAS). https://doi.org/10.1109/ICORAS.2017.8308068

10.1109/ICORAS.2017.830806828536344PMC5423489
19

Pulli K, Baksheev A, Kornyakov V, Eruhimov V (2012) Real-time computer vision with OpenCV. Commun ACM 55:61-69. https://doi.org/10.1145/2184319.2184337

10.1145/2184319.2184337
20

Rehman TU, Mahmud MS, Chang YK, Jin J, Shin J (2019) Current and future applications of statistical machine learning algorithms for agricultural machine vision systems. Comput Electron Agric 156:585-605. https://doi.org/10.1016/j.compag.2018.12.006

10.1016/j.compag.2018.12.006
21

Shahapure KR, Nicholas C (2020) Cluster quality analysis using silhouette score. Paper presented at the 2020 IEEE 7th International Conference on Data Science and Advanced Analytics (DSAA). https://doi.org/10.1109/DSAA49011.2020.00096

10.1109/DSAA49011.2020.00096
22

Shao M, Liu W, Zha L, Zhou C, Zhang Y, Li B (2020) Differential effects of high light duration on growth, nutritional quality, and oxidative stress of hydroponic lettuce under red and blue LED irradiation. Sci Hortic 268:109366. https://doi.org/10.1016/j.scienta.2020.109366

10.1016/j.scienta.2020.109366
23

Simko I, Hayes RJ, Furbank RT (2016) Non-destructive phenotyping of lettuce plants in early stages of development with optical sensors. Front Plant Sci 7:1985. https://doi.org/10.3389/fpls.2016.01985

10.3389/fpls.2016.0198528083011PMC5187177
24

Sinaga KP, Yang MS (2020) Unsupervised K-means clustering algorithm. IEEE Access 8:80716-80727. https://doi.org/10.1109/ACCESS.2020.2988796

10.1109/ACCESS.2020.2988796
25

Steidle Neto AJ, Moura LO, Lopes DC, Carlos LA, Martins LM, Ferraz LCL (2017) Non-destructive prediction of pigment content in lettuce based on visible-NIR spectroscopy. J Sci Food Agric 97:2015-2022. https://doi.org/10.1002/jsfa.8002

10.1002/jsfa.800227553517
26

Tan WH, Ibrahim H, Chan DJC (2021) Estimation of mass, chlorophylls, and anthocyanins of Spirodela polyrhiza with smartphone acquired images. Comput Electron Agric 190:106449. https://doi.org/10.1016/j.compag.2021.106449

10.1016/j.compag.2021.106449
27

Thomas S, Kuska MT, Bohnenkamp D, Brugger A, Alisaac E, Wahabzada M, Behmann J, Mahlein AK (2018) Benefits of hyperspectral imaging for plant disease detection and plant protection: A technical perspective. J Plant Dis Prot 125:5-20. https://doi.org/10.1007/s41348-017-0124-6

10.1007/s41348-017-0124-6
28

Tian K, Li J, Zeng J, Evans A, Zhang L (2019) Segmentation of tomato leaf images based on adaptive clustering number of K-means algorithm. Comput Electron Agric 165:104962. https://doi.org/10.1016/j.compag.2019.104962

10.1016/j.compag.2019.104962
29

Wang YM, Li Y, Zheng JB (2010) A camera calibration technique based on OpenCV. Paper presented at the The 3rd International Conference on Information Sciences and Interaction Sciences. https://doi.org/10.1109/ICICIS.2010.5534797

10.1109/ICICIS.2010.5534797
30

Xue L, Yang L (2009) Deriving leaf chlorophyll content of green-leafy vegetables from hyperspectral reflectance. ISPRS J Photogramm Remote Sens 64:97-106. https://doi.org/10.1016/j.isprsjprs.2008.06.002

10.1016/j.isprsjprs.2008.06.002
31

Yoo KO, Jang SW (2003) Intraspecific relationships of Lactuca sativa var. capitata cultivars based on RAPD analysis. Korean J Hort Sci Technol 21:273-278.

Information
  • Publisher :KOREAN SOCIETY FOR HORTICULTURAL SCIENCE
  • Publisher(Ko) :한국원예학회
  • Journal Title :Horticultural Science and Technology
  • Journal Title(Ko) :원예과학기술지
  • Received Date : 2023-09-20
  • Revised Date : 2024-03-20
  • Accepted Date : 2024-03-21
  • DOI :https://doi.org/10.7235/HORT.20250005
Journal Informaiton Horticultural Science and Technology Horticultural Science and Technology
Online Submission
  • scopus
  • NRF
  • KOFST
  • crossref crossmark
  • crossref cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close