Korean Journal of Agricultural Science (농업과학연구)

Institute of Agricultural Science, CNU (충남대학교 농업과학연구소)
권호 표지
DOI QR코드

DOI QR Code

Potential of multispectral imaging for maturity classification and recognition of oriental melon

  • Seongmin Lee (Department of Agricultural Engineering, National Institute of Agricultural Sciences) ;
  • Kyoung-Chul Kim (Department of Agricultural Engineering, National Institute of Agricultural Sciences) ;
  • Kangjin Lee (Department of Agricultural Engineering, National Institute of Agricultural Sciences) ;
  • Jinhwan Ryu (Department of Agricultural Engineering, National Institute of Agricultural Sciences) ;
  • Youngki Hong (Department of Agricultural Engineering, National Institute of Agricultural Sciences) ;
  • Byeong-Hyo Cho (Department of Agricultural Engineering, National Institute of Agricultural Sciences)
  • Received : 2023.07.28
  • Accepted : 2023.08.29
  • Published : 2023.09.01
Download PDF
⟨ Previous Next ⟩

Abstract

In this study, we aimed to apply multispectral imaging (713 - 920 nm, 10 bands) for maturity classification and recognition of oriental melons grown in hydroponic greenhouses. A total of 20 oriental melons were selected, and time series multispectral imaging of oriental melons was 7 - 9 times for each sample from April 21, 2023, to May 12, 2023. We used several approaches, such as Savitzky-Golay (SG), standard normal variate (SNV), and Combination of SG and SNV (SG + SNV), for pre-processing the multispectral data. As a result, 713 - 759 nm bands were preprocessed with SG for the maturity classification of oriental melons. Additionally, a Light Gradient Boosting Machine (LightGBM) was used to train the recognition model for oriental melon. R2 of recognition model were 0.92, 0.91 for the training and validation sets, respectively, and the F-scores were 96.6 and 79.4% for the training and testing sets, respectively. Therefore, multispectral imaging in the range of 713 - 920 nm can be used to classify oriental melons maturity and recognize their fruits.

Keywords

Acknowledgement

본 연구는 농림축산식품부 및 과학기술정보통신부, 농촌진흥청의 재원으로 농림식품기술기획평가원과 재단법인 스마트팜 연구개발사업단의 스마트팜 다부처패키지 혁신기술개발사업(421031-04)의 지원을 받아 연구되었습니다.

References

  1. Abdulhammed R, Musafer H, Alessa A, Faezipour M, Abuzneid A. 2019. Features dimensionality reduction approaches for machine learning based network intrusion detection. Electronics 8:322. DOI:10.3390/electronics8030322.
  2. Aggarwal SLP, Papay FA. 2022. Applications of multispectral and hyperspectral imaging in dermatology. Experimental Dermatology 31:1128-1135. DOI:10.1111/exd.14624.
  3. Arink M, Khan HA, Polder G. 2021. Light penetration properties of visible and NIR radiation in tomatoes applied to nondestructive quality assessment. Engineering Proceedings 9:18. DOI:10.3390/engproc2021009018.
  4. ATC (Agricultural Technology Center). 2023. Cultivation status of oriental melon in South Korea. Accessed in https://www.sj.go.kr/sj-atc/main.do on 11 July 2023. [in Korean]
  5. Baek Y, Kim HY. 2018. ModAugNet: A new forecasting framework for stock market index value with an overfitting prevention LSTM module. Expert System with Application 113:457-480. DOI:10.1016/j.eswa.2018.07.019.
  6. Benelli A, Cevoli C, Ragni L, Fabbri A. 2021. In-field and non-destructive monitoring of grapes maturity by hyperspectral imaging. Biosystems Engineering 207:59-67. DOI:10.1016/j.biosystemseng.2021.04.006.
  7. Cen H, He Y. 2007. Theory and application of near infrared reflectance spectroscopy in determination of food quality. Trends in Food Science & Technology 18:72-83. DOI:10.1016/j.tifs.2006.09.003.
  8. Choi YJ, Chun H, Lee SY, Kim HJ, Choi HY, Shin YS. 2007. Distribution of β-carotene in greenhouse oriental melon (Cucumis melo L.) fruit and its content following different ripening stage. pp. 185-189. In Proceeding of The Korean Society for Bio-Environment Control. [in Korean]
  9. Chun P, Izumi S, Yamane T. 2021. Automatic detection method of cracks from concrete surface imagery using two-step Light Gradient Boosting Machine. Computer-Aided Civil and Infrastructure Engineering 36:61-72. DOI:10:1111/mice.12564. 10:1111/mice.12564
  10. ElMasry G, Mandour N, Al-Rejaie S, Belin E, Rousseau D. 2019. Recent applications of multispectral imaging in seed phenotyping and quality monitoring-An overview. Sensors 19:1090. DOI:10.3390/s19051090.
  11. Fountas S, Mylonas N, Malounas I, Rodias E, Hellmann Santos C, Pekkeriet E. 2020. Agricultural robotics for field operations. Sensors 20:2672. DOI:10.3390/s20092672.
  12. Gholamy A, Kreinovich V, Kosheleva O. 2018. Why 70/30 or 80/20 relation between training and testing sets: A pedagogical explanation. International Journal of Intelligent Technologies and Applied Statistics 11:105-111. DOI:10.6148/Ijitas.201806_11(2).0003.
  13. Ha YS, Kim TW. 2014. Development of oriental melon harvesting robot in greenhouse cultivation. Journal of BioEnvironment Control 23:123-130. DOI:10.12791/KSBEC.2014.23.2.123. [in Korean]
  14. Hao X, Zhang Z, Huang G, Wang K. 2022. Prediction of f-CaO coment clinker: A novel prediction method based on LightGBM and Bayesian optimization. Chemometrics and Intelligent Laboratory Systems 220:104461. DOI:10.1016/j.chemolab.2021.104461.
  15. Hondo T, Kobayashi K, Aoyagi Y. 2022. Real-time prediction of growth characteristics for individual fruits using deep learning. Sensors 22:6473. DOI:10.3390/s22176473.
  16. Jia B, Wang W, Ni X, Lawrence KC, Zhuang H, Yoon SC, Gao Z. 2020. Essential processing methods of hyperspectral images of agricultural and food products. Chemometrics and Intelligent Laboratory Systems 198:103936. DOI:10.1016/j.chemolab.2020.103936.
  17. Jiang H, Jiang X, Ru Y, Chen Q, Li X, Xu L, Zhou H, Shi M. 2022. Rapid and non-destructive detection of natural mildew degree of postharvest Camellia oleifera fruit based on hyperspectral imaging. Infrared Physics & Technology 123:104169. DOI:10.1016/j.infrared.2022.104169.
  18. Kang Z, Zhao Y, Chen L, Guo Y, Mu Q, Wang S. 2022. Advances in machine learning and hyperspectral imaging in the food supply chain. Food Engineering Reviews 14:596-616. DOI:10.1007/s12393-022-09322-2.
  19. Kosec M, Burmen M, Tomazevic D, Pernus F, Likar B. 2013. Characterization of a spectrograph based hyperspectral imaging system. Optics Express 21:12085-12099. DOI:10.1364/OE.21.012085.
  20. KOSIS (Korean Statistical Information Service). 2023. Agriculture and forestry. Accessed in https://kosis.kr on 11 July 2023. [in Korean]
  21. Lee DS, Kwon JK, Yoon SW, Lee SY, Seo MT, Lee HJ, Lee SG, Kang TG. 2021. Comparison of yield and work intensity according to stem attraction method in melon hydroponics. Journal of Bioenvironmental Control 30:377-382. DOI:10.12791/KSBEC.2021.30.4.377. [in Korean]
  22. Li X, Wei Y, Xu J, Feng X, Wu F, Zhou R, Jin J, Xu K, Yu K, Yu X, et al. 2018. SSC and pH for sweet assessment and maturity classification f harvested cherry fruit based on NIR hyperspectral imaging technology. Postharvest Biology and Technology 143:112-118. DOI:10.1016/j.postharvbio.2018.05.003.
  23. Lowe A, Harrison N, French AP. 2017. Hyperspectral image analysis techniques for the detection and classification of the early onset of plant disease and stress. Plant Methods 13:80. DOI:10.1186/s13007-017-0233-z.
  24. Luo J, Bai J. 2005. Savitzky-Golay smoothing and differentiation filter for even number data. Signal Process 85:1429-1434. DOI:10.1016/j.sigpro.2005.02.002.
  25. Manea D, Calin MA. 2015. Hyperspectral imaging in different light conditions. The Imaging Science Journal 63:214-219. DOI:10.1179/1743131X15Y.0000000001.
  26. Massaoudi M, Refaat SS, Abu-Rub H, Chihi I, Oueslati FS. 2020. PLS-CNN-BiLSTM: An end-to-end algorithm-based Savitzky-Golay smoothing and evolution strategy for load forecasting. Energies 13:5464. DOI:10.3390/en13205464.
  27. Nguyen HDD, Pan V, Pham C, Valdez R, Doan K, Nansen C. 2020. Night-based hyperspectral imaging to study association of horticultural crop leaf reflectance and nutrient status. Computers and Electronics in Agriculture 173:105458. DOI:10.1016/j.compag.2020.105458.
  28. Roy IG. 2020. An optimal Savitzky-Golay derivative filter with geophysical applications: An example of self-potential data. Geophysical Prospecting 68:1041-1056. DOI:10.1111/1365-2478.12892.
  29. Rustam F, Reshi AA, Mehmood A, Ullah S, On BW, Aslam W, Choi GS. 2020. COVID-19 future forecasting using supervised machine learning models. IEEE access 8:101489-101499. DOI:10.1109/ACCESS.2020.2997311.
  30. Schober P, Boer C, Schwarte LA. 2018. Correlation coefficients: Appropriate use and interpretation. Anesthesia & Analgesia 126:1763-1768. DOI:10.1213/ANE.0000000000002864.
  31. Sun X, Liu M, Sima Z. 2020. A novel cryptocurrency price trend forecasting model based on lightGBM. Finance Research Letters 32:101084. DOI:10.1016/j.frl.2018.12.032.
  32. Taghinezhad E, Szumny A, Figiel A. 2023. The application of hyperspectral imaging technologies for the prediction and measurement of the moisture content of various agricultural crops during the drying process. Molecules 28:2930. DOI:10.3390/molecules28072930.
  33. Wang L, Sole A, Hardeberg JY. 2022a. Densely residual network with dual attention for hyperspectral reconstruction from RGB images. Remote Sensing 14:3128. DOI:10.3390/rs14133128.
  34. Wang M, Xu Y, Mo B, Nikitina MA, Xiao X. 2022b. Vis/NIR optical biosensors applications for fruit monitoring. Biosensors and Bioelectronics: X 11:100197. DOI:10.1016/j.biosx.2022.100197.
  35. Xie C, Chu B, He Y. 2018. Prediction of banana color and firmness using a novel wavelengths selection method of hyperspectral imaging. Food Chemistry 245:132-140. DOI:10.1016/j.foodchem.2017.10.079.
  36. Xu JT, Zhang Z, Guo YH, Liu YF, Ning XF. 2023. Detection of cucumber powdery mildew based on spectral and image information. Journal of Biosystems Engineering 48:115-122. DOI:10.1007/s42853-023-00178-w.
  37. Yang W, Xiong Y, Xu Z, Li L, Du Y. 2022. Piecewise preprocessing of near-infrared spectra for improving prediction ability of a PLS model. Infrared Physics & Technology 126:104359. DOI:10.1016/j.infrared.2022.104359.
  38. Zeb A, Qureshi WS, Ghafoor A, Malik A, Imran M, Imran M, Iqbal J, Alanazi E. 2021. Is this melon sweet? A quantitative classification for near-infrared spectroscopy. Infrared Physics & Technology 114:103645. DOI:10.1016/j.infrared.2021.103645.
  39. Zude M. 2003. Comparison of indices and multivariate models to non-destructively predict the fruit chlorophyll by means of visible spectrometry in apple fruit. Analytica Chimica Acta 481:119-126. DOI:10.1016/S0003-2670(03)00070-9.