Please use this identifier to cite or link to this item:
標題: 基於氣候因子之水稻稻熱病分類模型研究
Rice Blast Disease Classification based on Weather Factors
作者: 林家頡 
Chia-Chieh Lin 
關鍵字: 機器學習;稻熱病;Machine Learning;Rice Blast
引用: [1] M. Feurer, A. Klein, K. Eggensperger, J. Springenberg, M. Blum, and F. Hut- ter, 'Efficient and robust automated machine learning,' in Advances in Neural Information Processing Systems, pp. 2962–2970, 2015. [2] 蔡武雄, '行政院農委會動植物防疫檢疫局-植物保護圖鑑-稻熱病,' [3] P. Rini, M. Dipankar, B. Naik, et al., 'Effect of different meteorological param- eters on the development and progression of rice leaf blast disease in western odisha.,' International Journal of Plant Protection, vol. 10, no. 1, pp. 52–57, 2017. [4] Y. Kim, J.-H. Roh, and H. Y. Kim, 'Early forecasting of rice blast disease using long short-term memory recurrent neural networks,' Sustainability, vol. 10, no. 1, p. 34, 2017. [5] A. R. Malicdem and P. L. Fernandez, 'Rice blast disease forecasting for northern philippines,' WSEAS Trans. Inf. Sci. Appl, vol. 12, pp. 120–129, 2015. [6] I. Guyon and A. Elisseeff, 'An introduction to variable and feature selection,' Journal of machine learning research, vol. 3, no. Mar, pp. 1157–1182, 2003. [7] S. Chakraborty, R. Ghosh, M. Ghosh, C. D. Fernandes, M. Charchar, and S. Kelemu, 'Weather-based prediction of anthracnose severity using artificial neural network models,' Plant Pathology, vol. 53, no. 4, pp. 375–386, 2004. [8] K. Klem, M. Vanova, J. Hajslova, K. Lancová, and M. Sehnalová, 'A neural network model for prediction of deoxynivalenol content in wheat grain based on weather data and preceding crop,' Plant Soil and Environment, vol. 53, no. 10, p. 421, 2007.
水稻稻熱病為台灣水稻生長之主要疾病,只要氣候因素符合病菌生長條件,其病菌散布快速,約二至三天可導致大規模災害並造成大規模流行,染病嚴重會導致稻作死亡,造成嚴重農業損害,因此我們希望藉由歷史的氣候資料與葉稻熱病資料,以機器學習技術建構一個分類模型,預測當前的氣候因子是否可能導致稻熱病情加重,進而提早預防,降低農業損失。研究中所使用的資料集是嘉義大學植物醫學系與行政院農委會防檢局所提供,資料包含103、104、105、106、107五個年度氣候因子與稻熱病發病機率資料,我們篩選出了地理位置相近的雲林嘉義地區資料作為研究資料。因葉稻熱病是最常見之病狀,在每個生長期皆可能出現,農委會所提供之稻熱病發病率資料也以葉稻熱病為主,所以研究以對葉稻熱病的預測為主軸。我們依據稻熱病發病誘因挑選特徵值,並在實驗中以Recursive Feature Elimination演算法分析影響葉稻熱病發病的重要特徵,以Auto-Sklearn與神經網路建立出分類器。透過特徵分析與特徵集的調整,我們所建立出來的分類器有72%左右的準確率可以正確預測稻熱病情可能加重或減輕,而稻熱病情加重的情況能有89%的機率被分類器預測出。

Rice blast disease is the disease that has the greatest impact on rice growth in Taiwan. The goal of this research is to correlate historical climate data and rice blast disease data by classification models to predict if the rice blast disease will be exacerbated by a given climatic conduction. The data set used in the study was provided by the council of agriculture, executive yuan, Taiwan. The data we use are five annual climatic data (ranging from 2014 to 2018) and the field observation of rice blast disease during these years. With the data, we conduct feature selection by recursive feature elimination algorithm to analyze the key features on the rice blast disease. Through the features correlation analysis, we learn classification models by auto-sklearn and neural network. The experiment result shows that our model is with an accuracy of about 72% to correctly predict the condition (exacerbated or relived) of rice blast diseases. For the exacerbation case, our model can have 89%accuracy, demonstrating the effectiveness of the proposed classification model.
Rights: 同意授權瀏覽/列印電子全文服務,2019-01-30起公開。
Appears in Collections:資訊科學與工程學系所

Files in This Item:
File SizeFormat Existing users please Login
nchu-108-5103056020-1.pdf1.74 MBAdobe PDFThis file is only available in the university internal network    Request a copy
Show full item record

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.