稻熱病是水稻種植國家的主要傳染病之一，氣候適合就會發生且蔓延相當迅速，疾病防治對策可謂相當重要。近期的商用無人機具有靈活性高、操作簡易等優勢，搭載了攝影機的無人機可以拍攝相當大的面積。倘若搭載拍攝波段包含近紅外光的多光譜相機，則能以不同於可見光影像的角度來觀測物件。
　　本研究的動機在於觀察到傳統農家應對稻熱病的方法是以頻繁巡田來監測水稻狀態，但實際上不僅耗費人力而且需要大量相關知識。為了改善這種狀況，本論文設計出一套針對稻熱病的檢測系統。我們研究多光譜影像中水稻與其他地表物件的反射特徵，以類神經網路來分辨水稻葉與其他物件的影像位置還有對病況嚴重程度做分級，再使用群集分析來搜尋是否有群聚感染的狀況。
　　以本次觀測的後壁區水稻田為例，本論文製作的模型能夠清楚分辨水稻葉與其他非水稻物件，且能對葉稻熱病感染嚴重程度做分級，而後的OPTICS演算法則能找出疑似感染的群集。

Rice blast is one of the serious diseases in rice production countries. It spread rapidly when the climate is suitable. Therefore, prevention and control measures of the disease are considerable. UAV for commercial have the advantages of high flexibility and simplified operation these days, and it can shoot considerable measures of areas while equipped with cameras. If the multispectral camera has near-infrared band information, the object could be observed at an angle that differs from the visible image.
It has been observed that traditional farmers monitor rice blasts by a continuous patrol of the rice fields. However, it requires a lot of human resources and professional knowledge. In this thesis, a rice blast monitoring and forecasting model was proposed to resolve the problem. We studied the reflectance of rice and other surface objects in multispectral images. We used a neural network to recognized the location of rice leaf and other objects and classified the degree of the disease. Furthermore, this study applied cluster analysis to investigate the situation of cluster infections.
The model of the case studies on rice fields in the Houbi District in this thesis could distinctly distinguish the objects between rice leaves and others, and classified the severity of rice blast infection. Additionally, the OPTICS algorithm could identify the suspected infection clusters.

第一章　緒論 1
第二章　無人機與航空攝影 5
第三章　稻熱病與光譜特性 11
第四章　類神經網路 27
第五章　群集分析 33
第六章　實驗架構與結果 41
第七章　結論與未來展望 85
參考文獻 89

[1]108年臺閩地區農產品生產量值年報，檢自https://agrstat.coa.gov.tw/sdweb/public/official/OfficialInformation.aspx (Jan. 02, 2021)
[2]鄭萬助，107年我國糧食供需統計結果，農政與農情318期，頁 100-103，民國 107 年 12 月。
[3]A.A. Gitelson, M.N. Merzlyak, H.K. Lichtenthaler, Detection of red edge position and chlorophyll content by reflectance measurements near 700 nm, Journal of Plant Physiology Volume 148, pp. 501–508, 1996.
[4]D.N.H.Horler, M.Dockray, J.Barbera, A.R.Barringer, Red edge measurements for remotely sensing plant chlorophyll content, Advances in Space Research Volume 3, pp. 273–277, 1983.
[5]I. Filella, J. Penuelas, The red edge position and shape as indicators of plant chlorophyll content, biomass and hydric status, International Journal of Remote Sensing Volume 15, pp. 1459–1470, 1994.
[6]杜亮呈，無人飛行載具多光譜影像正射及融合之可行性暨精度評估，碩士論文，國立臺北科技大學土木工程系土木與防災碩士班，2016。
[7]楊宗祐，自製紅外線相機於無人載具影像之精度評估及應用，碩士論文，國立臺北科技大學土木工程系土木與防災碩士班，2016。
[8]維基百科Unmanned aerial vehicle。檢自https://en.wikipedia.org/wiki/Unmanned_aerial_vehicle (Oct. 09, 2020)
[9]無人飛行載具系統介紹。https://www.nlsc.gov.tw/UAS/2-1_UASintroduce.html (Oct. 09, 2020)
[10]DJI官方網站。檢自https://www.dji.com/tw/ (Oct. 09, 2020)
[11]林嘉宏，小型無人定翼機空拍路徑選擇評估之研究，國立高雄大學資訊管理學系，碩士論文，2014。
[12]Teppo Luukkonen, Modelling and control of quadcopter School of Science. Independent research project in applied mathematics, 2011.
[13]Wulder, M.A., Franklin, S.E., Remote sensing of forest environments: concepts and case studies, 2003.
[14]Factors Affecting Image Geometry。檢自https://nature.berkeley.edu/~penggong/textbook/chapter4/html/sect42.htm (Oct. 10, 2020)
[15]Lechne, A.M., Foody, G.M., Boyd, D.S., Applications in Remote Sensing to Forest Ecology and Management. One Earth 2020, 2, pp. 405–412.
[16]Radiometric Calibration and Corrections。檢自http://gsp.humboldt.edu/OLM/Courses/GSP_216_Online/lesson4-1/radiometric.html (Oct. 11, 2020)
[17]維基百科Vignetting。檢自https://en.wikipedia.org/wiki/Vignetting (Oct. 20, 2020)
[18] RedEdge Camera Radiometric Calibration Model。檢自https://support.micasense.com/hc/en-us/articles/115000351194-Rededge-Camera-Radiometric-Calibration-Model (Oct. 11, 2020)
[19]Janick F. Artiola, Ian L. Pepper, Mark L. Brusseau, Environmental Monitoring and Characterization, 2004.
[20]Lobell, D. B., Asner, G. P., Moisture effects on soil reflectance. Soil Science Society of America Journal. 66, pp. 722−727, 2002.
[21]Tian, J., Philpot, W.D., Relationship between surface soil water content, evaporation rate, and water absorption band depths in SWIR reflectance spectra. Remote Sens. Environ. 169, pp. 280−289, 2015.
[22]Sunlight。檢自https://en.wikipedia.org/wiki/Sunlight (Oct. 13, 2020)
[23]Photosynthesis。檢自https://en.wikipedia.org/wiki/Photosynthesis (Oct. 13, 2020)
[24]Photosynthetically active radiation。檢自https://en.wikipedia.org/wiki/Photosynthetically_active_radiation (Oct. 13, 2020)
[25]Inada, K., Spectral absorption property of pigments in the living leaves and its contribution to photosynthesis. Japanese Journal of Crop Science 49 (2), pp. 286–294, 1980.
[26] Gull, A., Lone, A.A., and Wani, N.U.I., Biotic and Abiotic Stresses in Plants, IntechOpen, 2019.
[27]HORLER, D. N. H., DOCKRAY, M., and BARBER, J., The red edge of plant leaf reflectance. International Journal of Remote Sensing, 4, pp. 273–288, 1983.
[28]Liu, Z.Y., Huang, J.F., Tao, R.X., Characterizing and estimating fungal disease severity of rice brown spot with hyperspectral reflectance data. Rice Sci. 15 (3), pp. 232–242, 2008.
[29]Das, P. K., B. Laxman, S. V. C. Kameswara Rao, M. V. R. Seshasai, and V. K. Dadhwal, Monitoring of bacterial leaf blight in rice using ground-based hyperspectral and LISS IV satellite data in Kurnool, Andhra Pradesh, India. International Journal of Pest Management 61(4), pp. 359–368, 2015.
[30]Rouse, J.W., Haas, R.H., Schell, J.A., and Deering, D.W., Monitoring Vegetation Systems in the Great Plains with ERTS, Proceedings of the Third Earth Resources Technology Satellite-1 Symposium, Greenbelt, NASA SP-351, vol. 1, pp. 309–317, 1973.
[31]Datt, B., 1998. Remote sensing of chlorophyll a, chlorophyll b, chlorophyll a b, and Total Carotenoid Content in Eucalyptus Leaves. Remote Sens. Environ. 66 (2), pp. 111–121, 1998.
[32]維基百科Normalized difference vegetation index。檢自https://en.wikipedia.org/wiki/Normalized_difference_vegetation_index (Jul. 30, 2020)
[33]What is NDVI。檢自https://gisgeography.com/ndvi-normalized-difference-vegetation-index/ (Jul. 30, 2020)
[34]作物病蟲害與肥培管理技術資料光碟。檢自https://web.tari.gov.tw/techcd/ (Jun. 13, 2019)
[35]植物保護圖鑑系列8 - 水稻保護。檢自https://www.baphiq.gov.tw/ws.php?id=4225 (Jun. 13, 2019)
[36]Dominik Scherer, Andreas C. Müller, and Sven Behnke, Evaluation of Pooling Operations in Convolutional Architectures for Object Recognition. In 20th International Conference Artificial Neural Networks (ICANN), pp. 92–101, 2010.
[37]Convolutional Neural Networks。檢自https://www.superdatascience.com/blogs/convolutional-neural-networks-cnn-summary/ (Sep. 20, 2020)
[38]Wim De Mulder, Steven Bethard, and Marie-francine Moens., A survey on the application of recurrent neural networks to statistical language modeling. Computer Speech & Language, 30(1), pp. 61–98, 2015.
[39]Kenji Suzuki., Artificial Neural Networks, Architectures and Applications, January 16th 2013.
[40]Artificial Neural Networks。檢自https://www.doc.ic.ac.uk/~nuric/teaching/imperial-college-machine-learning-neural-networks.html (Sep. 21, 2020)
[41]Leslie N. Smith and Nicholay Topin., Deep Convolutional Neural Network Design Patterns, 2016.
[42]Pytorch。檢自https://pytorch.org/ (Sep. 21, 2020)
[43] Yadira Q., Mario R., Carmen L., Iván T., Juan B., Autonomous Robot Navigation Based on Pattern Recognition Techniques and Artificial Neural Networks. in Bioinspired Computation in Artificial Systems, Springer International Publishing, 2015.
[44]維基百科Cluster analysis。檢自https://en.wikipedia.org/wiki/Cluster_analysis (Jan. 01, 2021)
[45]MacQueen, J. B., Some Methods for classification and Analysis of Multivariate Observations. 5th Berkeley Symposium on Mathematical Statistics and Probability, pp. 281–297, 1967.
[46] Cluster analysis。檢自https://datavizproject.com/data-type/cluster-analysis/ (Jan. 01, 2021)
[47]K-means Clustering with scikit-learn。檢自https://dev.to/nexttech/k-means-clustering-with-scikit-learn-14kk (Jan. 01, 2021)
[48] Martin Ester, Hans-Peter Kriegel, Jiirg Sander, Xiaowei Xu, A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise, Int. Conf. Knowledge Discovery and Data Mining, pp. 226–231, 1996.
[49] M. Ankerst, M. M. Breunig, H.-P. Kriegel, and J. Sander, OPTICS: ordering points to identify the clustering structure, in Proceedings of ACM SIGMOD Conference, pp. 49–60, June 1999.
[50]王涵平，後壁稻熱病嚴重全區快枯死 農民：目屎流袂離，檢自https://news.ltn.com.tw/news/life/breakingnews/2737976 (Aug. 02, 2020)
[51]王涵平，後壁稻熱病普遍發生 專家籲掌握防治時機，檢自https://news.ltn.com.tw/news/life/breakingnews/2740278 (Aug. 02, 2020)
[52]中央氣象局氣象資料開放平台。檢自https://opendata.cwb.gov.tw/index (Jun. 09, 2019)
[53]後壁區。檢自https://zh.wikipedia.org/wiki/%E5%BE%8C%E5%A3%81%E5%8D%80 (Aug. 20, 2020)
[54]林國清，水稻新品種台南11號之育成，台南區農業改良場研究彙報45期，頁 1-25，民國 94 年 6 月。
[55]RedEdge-M User Manual。檢自https://support.micasense.com/hc/en-us/articles/115003537673-RedEdge-M-User-Manual-PDF (Oct. 11, 2020)
[56]Kobayashi T, Kanda E, Kitada K, Ishiguro K, Torigoe Y, Detection of rice panicle blast with multispectral radiometer and the potential of using airborne multispectral scanners, Phytopathology 91, pp. 316–323, 2001.
[57]Kobayashi, T., Kanda, E., Naito, S. et al., Ratio of rice reflectance for estimating leaf blast severity with a multispectral radiometer. J Gen Plant Pathol 69, pp. 17–22, 2003.
[58]維基百科Confusion matrix。檢自https://en.wikipedia.org/wiki/Confusion_matrix (Nov. 01, 2020)
[59]維基百科Cohen’s kappa。檢自https://en.wikipedia.org/wiki/Cohen%27s_kappa (Nov. 01, 2020)
[60]維基百科卷積神經網路。檢自https://en.wikipedia.org/wiki/Convolutional_neural_network (Dec. 13, 2020)