한국농림기상학회지 (Korean Journal of Agricultural and Forest Meteorology)

  • 제21권4호
  • /
  • Pages.317-326
  • /
  • 2019
  • /
  • 1229-5671(pISSN)
  • /
  • 2288-1859(eISSN)
한국농림기상학회 (Korean Society of Agricultural and Forest Meteorology)
권호 표지
DOI QR코드

DOI QR Code

초분광 영상을 이용한 의사결정 트리 기반 봄감자(Solanum tuberosum)의 염해 판별

Application of Hyperspectral Imagery to Decision Tree Classifier for Assessment of Spring Potato (Solanum tuberosum) Damage by Salinity and Drought

  • 강경석 (애그로시스템공학전공 (농업생명과학원)) ;
  • 유찬석 (애그로시스템공학전공 (농업생명과학원)) ;
  • 장시형 (애그로시스템공학전공 (농업생명과학원)) ;
  • 강예성 (애그로시스템공학전공 (농업생명과학원)) ;
  • 전새롬 (애그로시스템공학전공 (농업생명과학원)) ;
  • 박준우 (애그로시스템공학전공 (농업생명과학원)) ;
  • 송혜영 (애그로시스템공학전공 (농업생명과학원)) ;
  • 이수환 (농촌진흥청 국립식량과학원)
  • Kang, Kyeong-Suk (Department of Agro-System Engineering, GyeongSang National University (Institute of Agriculture & Life Science)) ;
  • Ryu, Chan-Seok (Department of Agro-System Engineering, GyeongSang National University (Institute of Agriculture & Life Science)) ;
  • Jang, Si-Hyeong (Department of Agro-System Engineering, GyeongSang National University (Institute of Agriculture & Life Science)) ;
  • Kang, Ye-Seong (Department of Agro-System Engineering, GyeongSang National University (Institute of Agriculture & Life Science)) ;
  • Jun, Sae-Rom (Department of Agro-System Engineering, GyeongSang National University (Institute of Agriculture & Life Science)) ;
  • Park, Jun-Woo (Department of Agro-System Engineering, GyeongSang National University (Institute of Agriculture & Life Science)) ;
  • Song, Hye-Young (Department of Agro-System Engineering, GyeongSang National University (Institute of Agriculture & Life Science)) ;
  • Lee, Su Hwan (Nation Institute of Crop Science, Rural Development Administration)
  • 투고 : 2019.11.11
  • 심사 : 2019.12.04
  • 발행 : 2019.12.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 연구는 초분광 영상을 이용하여 간척지에서 주로 발생하는 염해 및 한해를 봄감자의 주요 생육단계에서 판별할 수 있는지를 검토하는 것이다. 영양생장기(VP), 괴경형성기(RFP) 및 괴경비대기(RGP)에 취득한 초분광 영상 내 봄감자 캐노피 영역의 반사율과 반사율의 불균일성을 최소화하기 위해 밴드 비로 변환하였다. 소형 다중분광 영상센서 개발을 고려하여 FWHM 5 nm의 단일 밴드를 상용화되어있는 밴드패스필터 기준으로 10 nm, 25 nm와 50 nm 평준화한 후 똑같이 밴드 비로 변화하였다. 의사결정트리법을 이용하여 각 FWHM에서 염해 판별에 유의한 단일 밴드 및 밴드 비를 추출하였고 그 분류 정확도는 OA와 KC로 나타내어졌다. 염해, 한해 및 정상 여부를 분류하기 위해 선택된 밴드는 최소 3개에서 최대 13개로 모든 FWHM에서 OA 66.7%와 KC 40.8% 이하의 정확도를 나타내었다. 괴경비대기(RGP)에서만 공통으로 440 nm가 선택되었고 동일 밴드는 아니지만 영양생장기(VP)에는 530 nm 또는 540 nm, 괴경비대기(RGP)에서는 추가로 710 nm 또는 720 nm가 선택되었다. 영양생장기(VP)에 비해 생식생장기(RFP 및 RGP)에 분류 정확도가 높지만 상용화가 용이한 10nm 이상의 FWHM에서 OA 및 KC값이 각각 78.7%, 57.7% 이하로 나타났다. 밴드 비를 이용하여 염해, 한해 및 정상을 분류하기 위해 선택된 밴드 비는 최소 2개에서 최대 6개로 원래 밴드(5 nm FWHM)의 비를 이용할 경우 생육 시기 및 FWHM에 관계없이 OA 및 KC가 95% 이상으로 나타났다. 영양생장기에서 FWHM에 관계없이 790 nm와 800 nm의 비가 선택되었고 동일 밴드는 아니지만 각 생육단계에서 Red, Red-edge 및 NIR 영역에서 유사밴드가 선택되었다. 모든 생육 시기에서 10 nm의 FWHM을 가진 3개 이하의 밴드 비를 이용한다면 OA 91.3%와 KC 85.0% 이상의 분류 정확도로 봄감자의 염해, 한해 및 정상여부판별이 가능할 것으로 판단된다. 이 결과는 넓은 면적에서 염해 및 한해 피해를 받은 작물 필지를 소형 다중 분광 카메라로 판별하여 빠르고 유연하게 제염기술을 투입하거나 그 피해 대책을 위한 정책 활용에 이용될 수 있을 것이다.

Salinity which is often detected on reclaimed land is a major detrimental factor to crop growth. It would be advantageous to develop an approach for assessment of salinity and drought damages using a non-destructive method in a large landfills area. The objective of this study was to examine applicability of the decision tree classifier using imagery for classifying for spring potatoes (Solanum tuberosum) damaged by salinity or drought at vegetation growth stages. We focused on comparing the accuracies of OA (Overall accuracy) and KC (Kappa coefficient) between the simple reflectance and the band ratios minimizing the effect on the light unevenness. Spectral merging based on the commercial band width with full width at half maximum (FWHM) such as 10 nm, 25 nm, and 50 nm was also considered to invent the multispectral image sensor. In the case of the classification based on original simple reflectance with 5 nm of FWHM, the selected bands ranged from 3-13 bands with the accuracy of less than 66.7% of OA and 40.8% of KC in all FWHMs. The maximum values of OA and KC values were 78.7% and 57.7%, respectively, with 10 nm of FWHM to classify salinity and drought damages of spring potato. When the classifier was built based on the band ratios, the accuracy was more than 95% of OA and KC regardless of growth stages and FWHMs. If the multispectral image sensor is made with the six bands (the ratios of three bands) with 10 nm of FWHM, it is possible to classify the damaged spring potato by salinity or drought using the reflectance of images with 91.3% of OA and 85.0% of KC.

키워드

참고문헌

  1. Bauriegel, E., A. Giebel, M. Geyer, U. Schmidt, and W. B. Herppich, 2011: Early detection of Fusarium infection in wheat using hyper-spectral imaging. Computers and Electronics in Agricultrue 75(2), 304-312. https://doi.org/10.1016/j.compag.2010.12.006
  2. Bresler. E., B. L. McNeal, and D. L. carter, 1982: Saline and sodic soils. Spring - Verlag Berlin Heidelberg New York, 167-171.
  3. Cawley, G. C., and N. L. Talbot, 2010: On over-fitting in model selection and subsequent selection bias in performance evaluation. Journal of Machine Learning Research 11(Jul), 2079-2107.
  4. Cho, J. W., K. S. Park, and S. Y. Lee, 2000: Growth characteristics of different barley cultivars grown on NaCl stress. Korean Journal of Environment Agriculture 19(1), 1-5.
  5. Collins, W., 1978: Remote sensing of crop type and maturity. Photogrammetric Engineering and Remote Sensing 44, 43-55.
  6. Dalal, R. C., and R. J. Henry, 1985: Simultaneous determination of moisture, organic carbon, and total Nitrogen by Near Infrared Reflectance spectrophotometry. SSSAJ 50(1), 120-123 https://doi.org/10.2136/sssaj1986.03615995005000010023x
  7. Goel, P. K., S. O. Prasher, R. M. Patel, J. A. Landry, R. B. Bonnell, and A. A. Viau, 2003: Classification of hyperspectral data by decision trees and artificial neural networks to identify weed stress and nitrogen status of corn. Computers and Electronics in Agriculture 39(2), 67-93. https://doi.org/10.1016/S0168-1699(03)00020-6
  8. Haboudane, D., J, R. Miller, E. Pattey, P. J. Zarco-Tejada, and I. B. Strachan, 2004: Hyperspectral vegetation indices and novel algorithms for predicting green LAI of crop canopies: Modeling and Validation in the Context of Precision Agriculture. Remote Sensing of Environment 90(3), 337-352. https://doi.org/10.1016/j.rse.2003.12.013
  9. Horler, D. N. H., M. Dockray, and J. Barber, 1983: The red edge of plant leaf reflectance. International Journal of Remote Sensing 4(2), 273-288. https://doi.org/10.1080/01431168308948546
  10. Kang, Y. S., C. S. Ryu, S. H. Kim, S. R. Kim, S. H. Jang, J. W. Park, T. K. Sarkar, and H. Y. Song, 2018: Yield prediction of Chinese cabbage (Brassicaceae) using broadband multispectral imagery mounted unmanned aerial system in the air and narrowband hyperspectral imagery on the ground. Journal of Biosystems Engineering 43(2), 138-147. https://doi.org/10.5307/JBE.2018.43.2.138
  11. Kim, S. H., K. S. Lee, J. R. Ma, and M. J. Kook, 2005: Currrent status of hyperspectral remote sensing: principle, data processing techniques, and applications. Korean Journal of Remote Sensing 21(4), 341-369. (in Korean with English abstract)
  12. Kim, Y., D. M. Glenn, J. Park, H. K. Ngugi, and B. L. Lehman, 2010: Hyperspectral image analysis for plant stress detection. In 2010 Pittsburgh, Pennsylvania, June 20-June 23, 2010 (p. 1). American Society of Agricultural and Biological Engineers.
  13. Lee, K. B., M. Xu, J. D. Kim, and K. Y. Jung, 2006: Soil characteristics and utilization on reclaimed land in Jangsu province coastal region of China. Journal of the Korean Society of International Agriculture.
  14. Lee, S. H., S. H. Yoo, S. I. Seol, Y. An, Y. S Jung, and S. M. Lee, 2000: Assessment of salt damage for upland-crop in Dae-Ho reclaimed soil. Korean Journal of Environment Agriculture 19(4), 358-363. (in Korean with English abstract)
  15. Lee, Y. H., H. S. Cho, J. H. Kim, W. G. Sang, P. Shin, J. K. Baek, and M. C. Seo, 2018: Effect of carbon dioxide concentration, temperature, and relative drought on growth responses and yield in spring potato (Solanum tuberosum. L). Korean Journal of Agricultural and Forest Meteorology 20(2), 149-158. (in Korean with English abstract) https://doi.org/10.5532/KJAFM.2018.20.2.149
  16. Mendenhall, M. J., and E. Merenyi, 2008: Relevance-based feature extraction for hyperspectral images. IEEE Transactions on Neural Networks 19(4), 658-672. https://doi.org/10.1109/TNN.2007.914156
  17. Martinez, J. L., H. Saulo, H. B. Escobar, and J. Leao, 2017: A new model selection criterion for partial least squares regression. Chemometrics and Intelligent Laboratory Systems 169, 64-78. https://doi.org/10.1016/j.chemolab.2017.08.006
  18. Mielikainen, J., 2006: Lossless compression of hyperspectral images using lookup tables. IEEE Signal Processing Letters 13(3), 157-16. https://doi.org/10.1109/LSP.2005.862604
  19. Pu, Y.Y., and D. W. Sun, 2016: Prediction of moisture content uniformity of microwave-vacuum dried Mangoes as affected by different shapes using NIR hyperspectral imaging. Innovative Food Science & Emerging Technologies 33, 348-356. https://doi.org/10.1016/j.ifset.2015.11.003
  20. Song, S., W. Gong, B. Zhu, and X. Huang, 2011: Wavelength selection and spectral discrimination for Paddy Rice with laboratory measurements of hyperspectral leaf reflectance. ISPRS Journal of Photogrammertry and Remote Sensing 66(5), 672-682. https://doi.org/10.1016/j.isprsjprs.2011.05.002
  21. Zarco-Tejada, P. J., V. Gonzalez-Dugo, and J. A. J. Berni, 2012: Fluorescence, temperature and narrow-band indices acquired from a UAV platform for water stress detection using a micro-hyperspectral imager and a Thermal Camera. Remote Sensing of Environment 117, 322-337. https://doi.org/10.1016/j.rse.2011.10.007