跳到主要內容

臺灣博碩士論文加值系統

(3.235.56.11) 您好!臺灣時間:2021/07/29 11:15
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:林尉喩
研究生(外文):Lin,Wei-Yu
論文名稱:發展以氣候資料為基礎的深度學習分類模式
論文名稱(外文):Developing a Deep Learning Classification Model Based on Climate Date
指導教授:李建邦李建邦引用關係
指導教授(外文):Lee,Chien-Pang
口試委員:林汶鑫黃淵科
口試委員(外文):Wen-Shin LinHuang, Yuan-Ko
口試日期:2020-07-24
學位類別:碩士
校院名稱:國立高雄科技大學
系所名稱:海事資訊科技系
學門:運輸服務學門
學類:航海學類
論文種類:學術論文
論文出版年:2020
畢業學年度:108
語文別:中文
論文頁數:45
中文關鍵詞:深度學習監督式學習分類演算法影像辨識
外文關鍵詞:Deep LearningSupervised LearningClassificationImage Recognition
相關次數:
  • 被引用被引用:0
  • 點閱點閱:51
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
近年來天氣異常,氣候變遷不僅是現在進行式,更有明顯加速的趨勢。許多國家開始出現氣候災難,例如:亞馬遜的野火、南北極圈的冰層消失、歐洲各國受熱浪侵襲,這樣極端氣候,將造成原有生態的改變與生態系統的破壞,甚至可能因此出現糧食危機進而影響到人類。
臺灣因為地理位置的關係,更易受氣候影響,為了降低氣候變遷對食物供給上的影響,本研究使用深度學習技術並在氣候因子的基礎下,設計一個預測稻熱病發生率的分類器。期望可讓農民使用與確認各地區的農作物未來稻熱病發生的機會,可以馬上防範,減少作物產量的損失。而本研究先以股票K線圖的概念將氣候資料轉換成圖形,再使用影像辨識方法中常用的Keras套件做建模找出特徵。依研究結果顯示,本研究平均結果較許多常見的分類方法來的好。在實驗四參數的設定下,本研究分類器建模效果最佳其平均正確率為93.3%。因此,未來值得以本研究之基礎和參數持續進行深入研究與探討。

Recently, the weather has been abnormal. Climate change is not only ongoing but also apparent acceleration. Many countries have many climate disasters. For example, wildfires in the Amazon, the disappearance of ice in the Arctic and Antarctic Circles, European countries are hit by heatwaves. The extreme climate would cause changes in the original ecology and destruction of the ecosystem and lead to a food crisis that would affect human survival.
Taiwan is more susceptible to climate influence due to its geographical location. In order to reduce the effect of climate change on food supply, this study uses deep learning. It is based on climate factors to develop a classification algorithm for classifying rice fever. Farmers would use the developed algorithm to confirm the occurrence of rice fever to prevent immediately and reduce the loss of crop yield. This study first uses the concept of stock candlestick chart to convert climate data into graphics. We then use the Keras, which is commonly used in image recognition issues, to build a classification model for searching features. According to the experiment results, the performance of this study is better than the other classification models. On the parameter settings of Experiment 4, this study brings the best average accuracy (93.3%). Therefore, it is worth continuing to conduct in-depth research and discussion on the basis and parameters of this study in the future.

摘 要………………………………………………………………ii
ABSTRACT………………………………………………………iii
誌謝………………………………………………………………iv
目錄………………………………………………………………v
表目錄……………………………………………………………vii
圖目錄……………………………………………………………ix
一. 前言……………………………………………………………1
1.1 研究背景………………………………………………………1
1.2 研究動機………………………………………………………1
1.3 研究目的………………………………………………………2
二. 文獻探討………………………………………………………3
2.1 氣候對臺灣產業之影響………………………………………3
2.2 氣候議題在分類方法之相關研究……………………………6
2.3 深度學習介紹…………………………………………………8
2.3.1 類神經網路…………………………………………………8
2.3.1.1 多層感知機(Multi Layer Perceptron)…………………9
2.3.1.2 徑向基底函式(Radial Basis Function)………………10
2.3.2 深度學習演算法…………………………………………11
2.3.2.1 卷積神經網路(Convolutional Neural Network)……12
2.3.2.2 循環神經網路(Recurrent Neural Network)…………14
2.4 常見深度學習平台…………………………………………15
2.4.1 TensorFlow………………………………………………16
2.4.2 PowerAI…………………………………………………16
2.4.3 CNTK……………………………………………………17
2.4.4 Caffe………………………………………………………18
2.4.5 Keras………………………………………………………19
三. 研究方法……………………………………………………21
3.1 資料前處理…………………………………………………21
3.2 繪圖…………………………………………………………24
3.3 分類方法……………………………………………………25
3.4 效能評估……………………………………………………26
3.5 比較方法……………………………………………………26
3.5.1 最近鄰演算法(K-Nearest Neighbor)……………………27
3.5.2 支援向量機(Support Vector Machine)…………………27
3.5.3 貝氏分類器(Bayesian Classifier)………………………28
3.5.4 決策樹(Decision Tree)……………………………………29
3.5.5 羅吉斯迴歸分析(Logistic regression)……………………30
四. 研究結果………………………………………………………31
4.1 資料來源與說明………………………………………………31
4.2 分類方法的建模效能比較與分析…………………………32
4.3 實驗結果與比較……………………………………………38
五. 結論與建議…………………………………………………42
參考文獻…………………………………………………………43

3SMARKET. (2018). 影響全球醫學界的11大AI事件(推薦閱讀). Retrieved from http://3smarket-info.blogspot.com/2018/01/11-ai.html
2.BBCNews中文. (2020). 肺炎疫情會引發食物危機嗎. Retrieved from https://www.bbc.com/zhongwen/trad/world-53263495
3.Breiman, L., Friedman, J. H., Olshen, R. A., & Stone, C. J. J. I. G. (1984). Classification and regression trees. Belmont, CA: Wadsworth. 432, 151-166.
4.Chang, Y., Lee, M.-A., Lee, K.-T., & Shao, K.-T. J. M. P. (2013). Adaptation of fisheries and mariculture management to extreme oceanic environmental changes and climate variability in Taiwan. 38, 476-482.
5.De Veaux, R. D., & Ungar, L. H. J. U. h. w. c. u. e. u. p. n.-i. p. (1997). A brief introduction to neural networks.
6.Fix, E., & Hodges Jr, J. (1951). Discriminatory analysis: nonparametric discrimination, consistency properties: USAF school of Aviation Medicine.
7.Gu, J., Lu, Z., Li, H., & Li, V. O. J. a. p. a. (2016). Incorporating copying mechanism in sequence-to-sequence learning.
8.Hopfield, J. J. J. P. o. t. n. a. o. s. (1982). Neural networks and physical systems with emergent collective computational abilities. 79(8), 2554-2558.
9.Hubel, D. H., & Wiesel, T. N. J. T. J. o. p. (1962). Receptive fields, binocular interaction and functional architecture in the cat's visual cortex. 160(1), 106.
10.Khan, T., Bertram, M. Y., Jina, R., Mash, B., Levitt, N., Hofman, K. J. D. r., & practice, c. (2013). Preventing diabetes blindness: cost effectiveness of a screening programme using digital non-mydriatic fundus photography for diabetic retinopathy in a primary health care setting in South Africa. 101(2), 170-176.
11.Lee, C.-P., Hsu, P.-Y., & Su, C.-C. J. S. o. T. T. E. (2019). Increased prevalence of Sjogren's syndrome in where soils contain high levels of chromium. 657, 1121-1126.
12.Lee, C.-P., Shieh, G.-J., Yiu, T.-J., & Kuo, B.-J. (2013). The strategy to simulate the cross-pollination rate for the co-existence of genetically modified (GM) and non-GM crops by using FPSOSVR. Chemometrics and Intelligent Laboratory Systems, 122, 50-57. doi:https://doi.org/10.1016/j.chemolab.2013.01.002
13.Long, E., Lin, H., Liu, Z., Wu, X., Wang, L., Jiang, J., . . . Chen, J. J. N. b. e. (2017). An artificial intelligence platform for the multihospital collaborative management of congenital cataracts. 1(2), 1-8.
14.Lynn. (2017). 類神經網路的復興:深度學習簡史. STOCKFEEL. Retrieved from https://www.stockfeel.com.tw/類神經網路的復興:深度學習簡史/
15.Minsky, M., & Papert, S. A. (2017). Perceptrons: An introduction to computational geometry: MIT press.
16.Powell, M. (2001). Radial basis function methods for interpolation to functions of many variables. Paper presented at the HERCMA.
17.Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1985). Learning internal representations by error propagation. Retrieved from
18.Rumelhart, D. E., Hinton, G. E., & Williams, R. J. J. n. (1986). Learning representations by back-propagating errors. 323(6088), 533-536.
19.Sak, H., Senior, A. W., & Beaufays, F. (2014). Long short-term memory recurrent neural network architectures for large scale acoustic modeling.
20.Yager, R. R. J. I. S. (2006). An extension of the naive Bayesian classifier. 176(5), 577-588.
21.Yan, L., Yoshua, B., & Geoffrey, H. J. n. (2015). Deep learning. 521(7553), 436-444.
22.交通部中央氣象局. (2020a). 每日雨量. Retrieved from https://www.cwb.gov.tw/V8/C/D/DailyPrecipitation.html
23.交通部中央氣象局. (2020b). 颱風資料庫. Retrieved from https://rdc28.cwb.gov.tw/TDB/public/warning_typhoon_list/
24.有機農業全球資訊網. (2018). 【庄頭開講】 極端氣候,農業風險誰承擔?20160708. Retrieved from https://info.organic.org.tw/3869/
25.行政院農業委員. (2010). 我國精準農業的發展方向與策略. Retrieved from https://www.coa.gov.tw/ws.php?id=2288&print=Y
26.行政院農業委員. (2019). 水稻葉稻熱病疫情現蹤 農委會協助農友減輕損失. Retrieved from https://www.coa.gov.tw/theme_data.php?theme=news&sub_theme=agri&id=7722
27.行政院農業委員. (2020). 107年我國糧食供需統計結果. Retrieved from https://www.coa.gov.tw/ws.php?id=2509988
28.李宗祐. (2017). 評估水文條件改變及河床-河水交互作用對七家灣溪河川流量與溪流棲地之影響. Retrieved from https://np.cpami.gov.tw/filesys/dl/type1/51/2819_09a77a2ad86799ba2cad06c1d6b42c1c.jpg
29.林家頡. (2019). 基於氣候因子之水稻稻熱病分類模型研究.
30.姚惠雯. (2020). 涼不涼? 天熱涼感衣銷量大增TVBS直擊成衣檢測室. TVBS. Retrieved from https://news.tvbs.com.tw/sports/1350838
31.洪珮瑜. (2020). 厚底、加長版雨鞋套百元價! 搜尋度增6倍. 東森新聞. Retrieved from https://news.ebc.net.tw/news/business/211030
32.莊哲權、周綾昀、謝佳潾. (2016). 颱風影響漁獲減 墾丁禁戲水. 中國時報. Retrieved from https://www.chinatimes.com/realtimenews/20161021000539-260503?chdtv
33.陳祐元. (2019). 颱風大雨頻攪局 蚵仔減產、白蝦死3成. TVBS新聞網. Retrieved from https://news.tvbs.com.tw/life/1188210
34.經濟部工業局產業節能減碳資訊網. (2015). 製造業氣候變遷調適指引. Retrieved from https://ghg.tgpf.org.tw/Counseling/counseling_sm7_7
35.蔡武雄. (2009). 水稻稻熱病研究回顧. 台灣水稻保護成果及新展望研討會專刊:, 1-12.

電子全文 電子全文(網際網路公開日期:20250827)
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top