Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography (한국측량학회지)

Korean Society of Surveying, Geodesy, Photogrammetry and Cartography (한국측량학회)
권호 표지
DOI QR코드

DOI QR Code

Development of a modified model for predicting cabbage yield based on soil properties using GIS

GIS를 이용한 토양정보 기반의 배추 생산량 예측 수정모델 개발

  • Received : 2022.10.13
  • Accepted : 2022.10.20
  • Published : 2022.10.31
Download PDF
⟨ Previous Next ⟩

Abstract

This study proposes a deep learning algorithm to predict crop yield using GIS (Geographic Information System) to extract soil properties from Soilgrids and soil suitability class maps. The proposed model modified the structure of a published CNN-RNN (Convolutional Neural Network-Recurrent Neural Network) based crop yield prediction model suitable for the domestic crop environment. The existing model has two characteristics. The first is that it replaces the original yield with the average yield of the year, and the second is that it trains the data of the predicted year. The new model uses the original field value to ensure accuracy, and the network structure has been improved so that it can train only with data prior to the year to be predicted. The proposed model predicted the yield per unit area of autumn cabbage for kimchi by region based on weather, soil, soil suitability classes, and yield data from 1980 to 2020. As a result of computing and predicting data for each of the four years from 2018 to 2021, the error amount for the test data set was about 10%, enabling accurate yield prediction, especially in regions with a large proportion of total yield. In addition, both the proposed model and the existing model show that the error gradually decreases as the number of years of training data increases, resulting in improved general-purpose performance as the number of training data increases.

본 연구는 GIS를 통해 토양정보를 수집하고 가공하여 농산물 생산량을 예측하는 모델을 제안한다. 농산물 생산량 예측 딥러닝 알고리즘은 공개된 CNN-RNN 농산물 생산량 예측 모델 구조를 변경하여 국내 농산물 자료 환경에 적합하도록 새롭게 구축하였다. 기존모델은 두 가지 특징을 가지고 있는데 첫 번째는 농산물의 생산량을 해당 필지값이 아닌 당해 평균값으로 대체한다는 것이고 두 번째는 예측하는 연도의 데이터까지 학습한다는 것이다. 새로운 모델은 해당 필지의 값을 그대로 사용하여 데이터의 정확성을 확보하고 예측하고자 하는 연도 이전의 데이터만 가지고 학습할 수 있도록 네트워크 구조를 개선하였다. 제안한 CNN-RNN 모델은 1980년부터 2020년까지의 기상정보, 토양정보, 토양적성도, 생산량 데이터를 학습하여 김장용 가을배추의 지역별 단위면적당 생산량을 예측한다. 2018년부터 2021년까지 4개 연도별 자료에 대하여 계산하고 생산량을 예측한 결과, 테스트 데이터셋에 대한 오차백분율이 약 10% 내외로 실제값과 비교하여 정확도 높은 생산량 예측이 가능했고, 특히 전체 생산량 비중이 큰 지역에서의 생산량은 비교적 근접하게 예측하는 것으로 분석되었다. 또한 제안모델과 기존모델은 모두 학습자료 연도 수가 증가할수록 점점 오차가 작아지므로 학습데이터가 많아질수록 범용 성능은 향상되는 결과를 나타낸다.

Keywords

Acknowledgement

본 연구는 중소벤처기업부의 기술혁신개발사업의 일환으로 수행하였음. [S2982642]

References

  1. Choi, S.C. (2016), Crop Yields Estimation Using Spatial Panel Regression Model, Master's thesis, Chonnam National University, Gwangju, Korea, 33p. (in Korean with English abstract)
  2. Kim, N., Ha, K.J., Park, N.W., Cho, J., Hong, S., and Lee, Y.W. (2019), A comparison between major artificial intelligence models for crop yield prediction: Case study of the midwestern united states, 2006-2015, ISPRS International Journal of Geo-Information, Vol. 8, 240. https://doi.org/10.3390/ijgi8050240
  3. Kim, J.H. and Kim, K.D. (2015), An outlook on chinese cabbage production by cultivation type under the RCP8.5 projected climate, Proceedings of The Korean Society of Agricultural and Forest Meteorology Conference-2015, 25 August, Jeonju, Korea, pp. 183-186. (in Korean with English abstract)
  4. Kim, S.W. and Kim, Y.h. (2021), A study on the application of machine learning algorithm to predict crop production, Journal of the Korea Academia-Industrial cooperation Society, Vol. 22, No. 7, pp. 403-408. (in Korean with English abstract) https://doi.org/10.5762/KAIS.2021.22.7.403
  5. Khaki, S., and Wang, L. (2019), Crop yield prediction using deep neural networks, Frontiers in Plant Science, Vol. 10, article 621. https://doi.org/10.3389/fpls.2019.00621
  6. Khaki, S., Wang, L. and Archontoulis, S. V, (2020), A CNNRNN framework for crop yield prediction, Frontiers in Plant Science, Vol. 10, article 1750. https://doi.org/10.3389/fpls.2019.01750
  7. Poggio L., de Sousa L.M., Batjes H.N., Heuvelink B.M.G., Kempen B., Ribeiro E., and Rossiter D., SoilGrids 2.0: producing soil information for the globe with quantified spatial uncertainty, SOIL, Vol. 7, pp. 217-240, 2021 https://doi.org/10.5194/soil-7-217-2021
  8. Lee, J.H., Lee, H.J., Kim, S.K., Lee, S.G., Lee, H.S., and Choi, C.S. (2017), Development of growth models as affected by cultivation season and transplanting date and estimation of prediction yield in kimch cabbage, Journal of BioEnvironment Control, Vol. 26, No. 4, pp. 235-241. (in Korean with English abstract) https://doi.org/10.12791/KSBEC.2017.26.4.235
  9. Lee, J.G. and Moon, A. (2015), Yield forecasting method for smart farming, Proceedings of the Korean Institute of Information and Commucation Sciences Conference-2015. 26 October, Busan, Korea, pp. 619-622. (in Korean with English abstract)
  10. National Honam Agricultural Experiment Station (2003), Soil Survey Theory and Practical Skills, National Honam Agricultural Experiment Station, National Institute of Agricultural Sciences, Jeollabuk-do Iksan (in Korean)