Korean Journal of Agricultural Science (농업과학연구)

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

DOI QR Code

Development of a classification model for tomato maturity using hyperspectral imagery

  • Hye-Young, Song (Department of Agricultural Engineering, National Institute of Agricultural Sciences) ;
  • Byeong-Hyo, Cho (Department of Agricultural Engineering, National Institute of Agricultural Sciences) ;
  • Yong-Hyun, Kim (Department of Agricultural Engineering, National Institute of Agricultural Sciences) ;
  • Kyoung-Chul, Kim (Department of Agricultural Engineering, National Institute of Agricultural Sciences)
  • Received : 2022.01.12
  • Accepted : 2022.02.28
  • Published : 2022.03.01
Download PDF
⟨ Previous Next ⟩

Abstract

In this study, we aimed to develop a maturity classification model for tomatoes using hyperspectral imaging in the range of 400 - 1,000 nm. Fifty-seven tomatoes harvested in August and November of 2021 were used as the sample set, and hyperspectral data was extracted from the surfaces of these tomatoes. A combined method of SNV (standard normal variate) and SG (Savitzky-Golay) methods was used for the pre-processing of the hyperspectral data. In addition, the hyperspectral data were analyzed for all maturity stages and considering bandwidths with different FWHM (full width at half maximum) values of 2, 25, and 50 nm. The PCA (principal component analysis) method was used to analyze the principal components related to maturity stages for the tomatoes. As a result, 500 - 550 nm and 650 - 700 nm bands were found to be related to the maturity stages of tomatoes. In addition, PC1 and PC2 explained approximately 97% of the variance at all FWHM conditions and thus were used as input data for classification model training based on the SVM (support vector machine). The SVM models were able to classify tomato maturity into five stages (Green, Turning, Pink, Light red, and Red) with over 95% accuracy regardless of the FWHM condition. Therefore, it was considered that hyperspectral data with 50 nm FWHM and SVM is feasible for use in the classification of tomato maturity into five stages.

Keywords

Acknowledgement

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

References

  1. Amigo JM, Santos C. 2020. Preprocessing of hyperspectral and multispectral images. Data Handling in Science and Technology 32:37-53. doi.org/10.1016/B978-0-444-63977-6.00003-1
  2. Benelli A, Cevoli C, Fabbri A, Ragni L. 2021. Ripeness evaluation of kiwifruit by hyperspectral imaging. Biosystems Engineering 2021. doi.org/10.1016/j.biosystemseng.2021.08.009
  3. Goel N, Sehgal P. 2015. Fuzzy classification of pre-harvest tomatoes for ripeness estimation-an approach based on automatic rule learning using decision tree. Applied Soft Computing 36:45-56. doi.org/10.1016/j.asoc.2015.07.009
  4. Guo C, Liu F, Kong W, He Y, Lou B. 2016. Hyperspectral imaging analysis for ripeness evaluation of strawberry with support vector machine. Journal of Food Engineering 179:11-18. doi.org/10.1016/j.jfoodeng.2016.01.002
  5. Hobson GE, Adams P, Dixon TJ. 1983. Assessing the colour of tomato fruit during ripening. Journal of the Science of Food and Agriculture 34:286-292. doi.org/10.1002/jsfa.2740340312
  6. Jensen JR. 2016. Introductory digital image processing: A remote sensing perspective 4th Edition. pp. 310-331. 5SIGMA PRESS, Seoul, Korea. [in Korean]
  7. Kang YS, Ryu CS, Jun SR, Jang SH, Park JW, Song HY, Lee WS. 2018. Distinguishing between closely related species of Allium and of Brassicaceae by narrowband hyperspectral imagery. Biosystems Engineering 176:103-113. doi.org/10.1016/j.biosystemseng.2018.10.003
  8. Khodabakhshian R, Emadi B. 2017. Application of Vis/SNIR hyperspectral imaging in ripeness classification of pear. International Journal of Food Properties 20:S3149-S3163. doi.org/10.1080/10942912.2017.1354022
  9. Kim DY, Cho BK, Kim YS. 2012. Non-destructive quality prediction of truss tomatoes using hyperspectral reflectance imagery. Korean Journal of Agricultural Science 39:413-420. [in Korean] doi.org/10.7744/cnujas.2012.39.3.413
  10. KOSIS (Korean Statistical Information Service). 2021. Agricultural area and crop production survey. Accessed in http://kosis.kr/index/index.do on 5 December 2021. [in Korean]
  11. Park SW, Lee JW, Kim YC, Kim KY, Hong SJ. 2004. Changes in fruit quality of tomato 'Dotaerang' cultivar during maturation and postharvest ripening. Horticultural Science and Technology 22:381-387. [in Korean]
  12. Polder G, Van der Heijden GWAM, Young IT. 2000. Hyperspectral image analysis for measuring ripeness of tomatoes. In Proceeding of ASABE Annual International Meeting, Paper No. 003089. doi.org/10.13031/2013.9924
  13. Rahman A, Park E, Bae H, Cho BK. 2018. Hyperspectral imaging technique to evaluate the firmness and the sweetness index of tomatoes. Korean Journal of Agricultural Science 45:823-837. doi.org/10.7744/KJOAS.20180075
  14. Rajkumar P, Wang N, EImasry G, Raghavan GSV, Gariepy Y. 2012. Studies on banana fruit quality and maturity stages using hyperspectral imaging. Journal of Food Engineering 108:194-200. doi.org/10.1016/j.jfoodeng.2011.05.002
  15. RDA (Rural Development Administration). 2012. RDA Interrobang. Accessed in https:// nongsaro.go.kr on 5 October 2021. [in Korean]
  16. Ruffin C, King RL. 1999. The analysis of hyperspectral data using Savitzky-Golay filtering-theoretical basis. 1. pp. 756-758. IEEE 1999 International Geoscience and Remote Sensing Symposium. IGARSS'99 (Cat. No. 99CH36293). doi.org/10.1109/IGARSS.1999.774430
  17. Savitzky A, Golay MJ. 1964. Smoothing and differentiation of data by simplified least squares procedures. Analytical Chemistry 36:1627-1639. doi.org/10.1021/ac60214a047
  18. Seo D, Cho BH, Kim KC. 2021a. Development of monitoring robot system for tomato fruits in hydroponic greenhouses. Agronomy 11:2211. doi.org/10.3390/agronomy11112211
  19. Seo Y, Mo C, Lim J, Lee A, Kim B, Jang J, Kim G. 2021b. Detection of spinach juice residues on stainless steel surfaces using VNIR hyperspectral images. Journal of Biosystems Engineering 46:173-181. doi.org/10.1007/s42853-021-00097-8
  20. Suh SR, Lee KH, Yu SH, Yoo SN, Choi YS. 2011. Comparison of performance of measuring method of VIS/NIR spectroscopic spectrum to predict soluble solids content of 'Shingo' pear. Journal of Biosystems Engineering 36:130-139. doi.org/10.5307/JBE.2011.36.2.130
  21. Wei X, Liu F, Qiu Z, Shao Y, He Y. 2014. Ripeness classification of astringent persimmon using hyperspectral imaging technique. Food and Bioprocess Technology 7:1371-1380. doi.org/10.1007/s11947-013-1164-y
  22. Yang C, Lee WS, Gader P. 2014. Hyperspectral band selection for detecting different blueberry fruit maturity stages. Computers and Electronics in Agriculture 109:23-31. doi.org/10.1016/j.compag.2014.08.009