기계학습 기반의 감귤 착과량 예측 시스템
A Prediction System for Amounts of Fruits of a Citrus Tree based on Machine Learning
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
농업기술에 정보통신기술(ICT)을 접목하기 시작하였으며, IoT(Internet of Things) 장치를 이용하여 측정된온도, 습도, 토양 등에 대한 데이터를 인공지능으로 분석하여 작물 생산을 제어하는 연구가 활발히 진행되고있다. 감귤은 생육단계에 대한 정보를 디지털 자료화할 수 있어서 정보통신기술을 접목한 생산제어가 가능하다. 본 논문에서는 감귤 착과량과 연관 관계가 있는 데이터를 이용하여 실제 감귤 착과량을 예측하는 새로운형태의 인공지능 기반 예측 방법을 제안한다. 제안하는 방법은 감귤 착과량을 예측할 수 있는 요인에 해당하는 데이터를 효과적으로 학습할 수 있는 예측 모델을 제시한다. 제시된 모델은 최적의 예측을 위하여 앙상블형태의 복합 모델로 구성된다. 학습된 예측 모델은 실제 측정한 제주 감귤 데이터를 이용한 예측 성능에 대한 실험을 수행하였다. 평가실험을 통하여 제안시스템이 효과적으로 감귤 생산량 예측을 할 수 있음을 보였다.
Abstract
Information and communication technology (ICT) is starting to be applied to agricultural technology. And Research is being actively conducted to control crop production by analyzing data measured using IoT devices with artificial intelligence technologies. Since information about the growth stages of Citrus can be converted into digital data, production control is possible by information and communication technology. In this paper, we propose an artificial intelligence method that predicts the actual amount of citrus. The proposed method presents a prediction model that can be effectively learn data corresponding to factors that can predict the amount of citrus. The model consists of a complex model in form of an ensemble for optimal prediction. The performance of the proposed model is verified using data obtained from actual citrus trees.
참고문헌 (10)
- S. Li, "INCOME : Practical land monitoring precision agriculture with sensor networks", Computer Communications, 36(4), pp.459~467 (2013)[DOI]
- S. Kim, "A Study on the Application of Machine Learning Algorithm to Predict Crop Production", Journal of the Korea Academia-Industrial cooperation Society, 22(7), pp.403~408 (2021)[DOI]
- A. Kaya, "Analysis of transfer learning for deep neural network based plant classification models", Computers and Electronics in Agriculture, 147(1), pp.70~90 (2019)[DOI]
- "Regulation and Pruning of citrus trees" (2014)[URL]
- D. Chung, "Evaluation of Forecasting Performance of Rice Yield Models under Climate Change", Korea Environmental Policy And Administration Society, 26(4), pp.197~222 (2018)[DOI]
- Oh, S., "Predicting Onion Production by Weather and Spatial Time Series Model", Journal of The Korean Data Analysis Society, 19(5), pp.2447~2456 (2017)
- Jiang, Z., "Predicting county-scale maize yields with publicly available data", Scientific Reports, 10(1) (2020)[DOI]
- Chen, T., "TADA : Trend Alignment with Dual-Attention Multi-Task Recurrent Neural Networks for Sales Prediction", Proceeding of IEEE International Conference on Data Mining, pp.49~58 (2018)
- "Citrus Fruit Amount Learning Data" (2022)[URL]
- "PyCaret: An open source, low-code machine learning library in Python" (2020)[URL]