臺灣博碩士論文加值系統

近年來，氣候變遷對於農業生產造成重大影響，如何維持農作物的產量儼然為農業領域上的一大課題。高溫以及降雨的改變使得植物的病害更為嚴重，而提前得知植物病害的發病狀況有助於避免病害擴散。由高光譜影像技術可以提前偵測到潛伏期的病徵，進而避免草莓炭疽病的蔓延。為了改善植物病害的辨識效率，本研究致力於建立一套手持式多光譜影像裝置以檢測植物病害。此裝置使用嵌入式系統當作系統的控制器，藉由放置於微型攝影機之前的濾鏡，可以擷取到所需要的特徵波段影像。藉由這些影像的資訊，可以得到不同波段的染病資訊。手持式多光譜裝置擷取四個波段的影像後，經過白校正影像處理以降低光線不均勻的影響後，藉由觀察接種炭疽病孢子液後的草莓葉片，由多光譜以及RGB影像資訊，可以分析不同時期的發病狀況。在本研究當中，我們首先以手持式多光譜裝置辨識草莓葉片的健康期以及病徵期兩種狀態，然後再進一步進行草莓葉片的健康期、潛伏期和病徵期三種狀態的辨識。在葉片平整的狀況下，本研究利用SVM模型在兩類病害分析上可以到達90.0%以上的準確率，而在三類病害分析上則可以在健康期、潛伏期以及染病期上面得到92.2%、68.6%和97.9%的準確率。染病草莓葉片以多光譜裝置取像後可以利用假彩色的方式呈現不同時期的病害症狀，讓使用者簡單且準確的得知染病植物的狀況，以採取適當防治措施。由於不平整葉片上的陰影會嚴重影響多光譜影像裝置的病害分析辨識率，因此我們進一步提出一個透過影像組合的方法來降低陰影所造成辨識率降低的影響。經過觀察多光譜影像彼此之間的關聯後，建立了四個較具有判別陰影、病徵以及健康區域的組合影像。以新的組合影像進行SVM訓練，健康期的辨識率由71.3%提高到95.7%，而病徵期的辨識率由82.3%提高到88.9%。

In recent years, the climate change has significantly affected the agricultural production. Maintaining the crop production is one of main concerns in agriculture. High temperature and changes of rainfall patterns enhance the spread of plant diseases. Hence it is desirable to seek for early detection of plant disease, and thus to control the spread of plant disease. Hyperspectral imaging has been proved to be an efficient tool for early detection of strawberry Anthracnose. To improve the efficiency of plant disease detection, this research aims to build a handheld multispectral imaging device for strawberry Anthracnose detection. This device uses an embedded system as the controller of the device. By placing filters in front of four miniature cameras, the images of four characteristic wavelengths are acquired. After capturing images using the handheld multispectral imaging device, images are processed to correct the effect of uneven lighting. Then by further processing the multispectral images and incorporating the RGB image of inoculated strawberry leaves, we are able to analyze the status of strawberry leaves at various infection stages. In this research, we first used the multispectral imaging device to classify the healthy and symptomatic areas in strawberry leaves. Then we further attempted to classify the status into three categories: healthy, incubation and symptomatic. SVM model was applied for classification of infection stages. For classification of healthy and symptomatic status, detection accuracy is above 90%. For classification between healthy, incubation, and symptomatic status, the accuracies are 92.2%, 68.6%, and 97.9%, respectively. The classification result of strawberry Anthracnose infection is further displayed on the handheld device as pseudo-color image so the user can easily observe the plant health condition, and so the disease management can be applied if necessary. Since the detection accuracy can be affected by lighting and shadow due to uneven surface of strawberry leaves. We propose a method to amend the effect of shadow on status classification. Through observations of the original four images and their association, a new set of images derived from the original four images was selected and tested to rectify the shadow effect. Using this new set of derived images and trained with SVM, classification accuracy for healthy status increased from 71.3% to 95.7% and the classification accuracy for symptomatic status also increased from 82.3% to 88.9%.

誌謝 i
摘要 iii
Abstract v
圖目錄 xi
表目錄 xv
第一章 緒論 1
1.1 前言 1
1.2 研究目的 3
第二章 文獻探討 5
2.1 植物病害 5
2.1.1 草莓炭疽病 5
2.2 植物病害檢測方法 6
2.2.1 植物病害檢測方法 6
2.2.2 前處理方法 6
2.2.3 特徵擷取方法 9
2.3 高光譜影像系統 10
2.3.1高光譜科技 10
2.3.2高光譜影像技術 10
2.3.3 高光譜影像檢測之應用 11
2.4 多光譜影像系統 13
2.5 分類方法 14
2.5.1 支持向量機 14
第三章 研究設備與方法 17
3.1 實驗材料 17
3.1.1 植物栽培 17
3.1.2 葉片前處理及固定 17
3.1.3接種處理 19
3.1.4後續保濕作業 20
3.2實驗器材 21
3.2.1拍攝彩色影像裝置 21
3.2.2 多光譜影像儀器使用器材 22
3.2.3 手持式多光譜影像儀器 27
3.2.4 多光譜拍攝裝置使用流程 27
3.3 手持式裝置硬體設計 28
3.3.1 手持式硬體設計 28
3.3.2 燈光電路 32
3.3.3 人機介面軟體設計 33
3.4 多光譜數據處理 35
3.4.1多光譜資料拍攝流程 36
3.4.2 資料前處理 40
3.4.3 資料對應與去除背景 46
3.4.4多光譜資料特徵點擷取 48
3.5 多光譜資料分類 50
3.5.1 兩類病害分析 50
3.5.2 三類病害分析 57
3.6 實驗樣本與分析樣本 58
第四章 結果與討論 61
4.1多光譜軟體數據處理跟分析 61
4.1.1白校正參考圖校正 61
4.1.2對位歪曲結果 68
4.1.3多光譜影像去背 70
4.1.4 多光譜圖片與RGB彩色圖對位 73
4.1.5特徵選點程式 76
4.1.6 病徵點判別 79
4.1.7建立SVM分類模型 80
4.2 二類病況分類效果 83
4.2.1 交叉驗證結果 83
4.2.2 假彩色結果 84
4.3 三類病況分類效果 88
4.3.1 炭疽病潛伏期辨識結果 88
4.3.2 留一驗證 (LOOCV) 結果 91
4.3.3 假彩色結果 91
4.4 陰影處理效果 96
4.4.1 前處理結果之陰影效果比較 96
4.4.2 前處理結果之假彩色效果 98
第五章 結論與建議 103
5.1 結論 103
5.2 建議 105
參考文獻 107

申雍、李裕娟. 2002. 簡易高光譜影像拍攝系統之研發與應用 應用於水稻精準農業體系之知識與技術 129-138: 行政院農業委員會農業試驗所.
李汪盛。2013。蕹菜葉綠素含量多光譜影像遙測系統之研發。桃園區農業改良場研究彙報 73: 37-52。
李窓明。2014。草莓本圃栽培管理。桃園區農業專訊 44: 15-20。
李聯興、葉俊巖、賴守正。1992。草莓主要病害綜合防治。農藥世界 104: 17-18。
林子傑。2010。農產品非破壞性檢測技術。台中區農業改良場特刊 105(1): 39-42。
柯建全、黃膺任、艾群。1999。應用影像處理檢測荔枝果皮之顏色變化。農業機械學刊 8(1): 59-68。
費雯綺、王喻其。 2007。 植物保護手冊-果樹篇 。 台中: 行政院農業委員會農業藥物毒物試驗所。
黃振文、彭玉湘。 2008。 有機農法的作物病害管理技術。出自〝有機作物栽培技術研討會專刊〞，13-24。劉大江、林俊義、簡宣裕、張明暉、劉禎祺主編。台中：農業試驗所。
黃益田、蕭淑芬。2011。全球氣候變遷對農作物病害發生影響及因應對策。氣候變遷對農作物生產之影響 71(1): 59-71。
鄭安秀、陳紹崇。 2000。 觀葉植物病害及防治。出自〝台南區農業改良場技術專刊〞，台南：台南區農業改良場。
鐘珮哲、彭淑貞、張廣淼、楊秀珠、余思葳。 2012。 草莓病蟲害之發生與管理 。 台北: 行政院農業委員會動植物防疫檢疫局、台中：行政院農業委員會農業藥物毒物試驗所。

Bazi, Y., and F. Melgani. 2006. Toward an Optimal SVM Classification System for Hyperspectral Remote Sensing Images. Geoscience and Remote Sensing, IEEE Transactions on, 44(11), 3374-3385. doi: 10.1109/TGRS.2006.880628
Blasco, J., N. Aleixos, J. Gómez, and E. Moltó. 2007. Citrus sorting by identification of the most common defects using multispectral computer vision. Journal of Food Engineering, 83(3), 384-393. doi: http://dx.doi.org/10.1016/j.jfoodeng.2007.03.027
Bradski, G. 2000. OpenCV Library. Dr. Dobb''s Journal of Software Tools.
Chang, C.-I. 2003. Hyperspectral Imaging: Techniques for Spectral Detection and Classification: Springer.
Chaudhary, P., A. K. Chaudhari, A. N. Cheeran, and S. Godara. 2012. Color Transform Based Approach for Disease Spot Detection on Plant Leaf. 3(6), 65-70.
Chung, W.-C. (2013). Detection and Analysis of Bok Choy Black Spot Disease and Strawberry Foliar Anthracnose Disease Using Hyperspectral Imaging. (Master).
Cortes, C., and V. Vapnik. 1995. Support-Vector Networks. Machine Learning, 20(3), 273-297.
Delwiche, S. R., and R. A. Graybosch. 2002. Identification of Waxy Wheat by Near-infrared Reflectance Spectroscopy. Journal of Cereal Science, 35(1), 29-38. doi: http://dx.doi.org/10.1006/jcrs.2001.0400
Geng, Y., M. Li, Y. Yuan, and Z. Hu. 2009. A Study on the Method of Image Pre-processing for Recognition of Crop Diseases. Paper presented at the Advanced Computer Control, 2009. ICACC ''09. International Conference on.
Gonzalez, R. C., and R. E. Woods. 2008. Intensity Transformations and Spatial Filtering Digital Image Processing 3 Edition 156-157: Prentice Hall.
Green, R. O., M. L. Eastwood, C. M. Sarture, T. G. Chrien, M. Aronsson, B. J. Chippendale, . . . O. Williams. 1998. Imaging Spectroscopy and the Airborne Visible/Infrared Imaging Spectrometer (AVIRIS). Remote Sensing of Environment, 65(3), 227-248. doi: http://dx.doi.org/10.1016/S0034-4257(98)00064-9
Howard, C. M., J. Maas, C. K. Chandler, and E. E. Albregts. 1992. Anthracnose of strawberry caused by the Colletotrichum complex in Florida. Florida Experiment Stations Journal Series No. R-01869, 76(10), 976-981.
Karimi, Y., S. O. Prasher, A. Madani, and S. Kim. 2008. Application of support vector machine technology for the estimation of crop biophysical parameters using aerial hyperspectral observations. Canadian Biosystems Engineering, 50(3), 13-20.
Marathe, H. D., and P. N. Kothe. 2013. Leaf Disease Detection Using Image Processing Technique. International Journal of Engineering Research & Technology (IJERT), 2(3).
Moshou, D., C. Bravo, R. Oberti, J. West, L. Bodria, A. McCartney, and H. Ramon. 2005. Plant disease detection based on data fusion of hyper-spectral and multi-spectral fluorescence imaging using Kohonen maps. Real-Time Imaging, 11(2), 75-83. doi: http://dx.doi.org/10.1016/j.rti.2005.03.003
Qin, J., T. F. Burks, M. S. Kim, K. Chao, and M. A. Ritenour. 2008. Citrus canker detection using hyperspectral reflectance imaging and PCA-based image classification method. Sensing and Instrumentation for Food Quality and Safety, 2(3), 168-177. doi: 10.1007/s11694-008-9043-3
Rumpf, T., A. K. Mahlein, U. Steiner, E. C. Oerke, H. W. Dehne, and L. Plümer. 2010. Early detection and classification of plant diseases with Support Vector Machines based on hyperspectral reflectance. Computers and Electronics in Agriculture, 74(1), 91-99. doi: http://dx.doi.org/10.1016/j.compag.2010.06.009
Sankaran, S., A. Mishra, R. Ehsani, and C. Davis. 2010. A review of advanced techniques for detecting plant diseases. Computers and Electronics in Agriculture, 72(1), 1-13. doi: http://dx.doi.org/10.1016/j.compag.2010.02.007
Shi, T., H. Liu, Y. Chen, J. Wang, and G. Wu. 2016. Estimation of arsenic in agricultural soils using hyperspectral vegetation indices of rice. Journal of Hazardous Materials, 308, 243-252. doi: http://dx.doi.org/10.1016/j.jhazmat.2016.01.022
Wang, L.-g., Y.-f. Gu, and Y. Zhang. 2005. Band Selection Method Based on Combination of Support Vector Machines and Subspatial Partition. China Academic Journal Electronic Publishing House. , 27(6).
Yeh, Y.-H. F., C.-L. Chung, J.-Y. Liao, C.-L. Chung, Y.-F. Kuo, and T.-T. Lin. 2013a. A comparison of machine learning methods on hyperspectral plant disease assessments. 2013 IFAC Bio-Robotics Conference. 1:(1)361-365
Yeh, Y.-H. F., W.-C. Chung, J.-Y. Liao, C.-L. Chung, Y.-F. Kuo, and T.-T. Lin. 2013b. 5th IFAC Conference on Bio-RoboticsA Comparison of Machine Learning Methods on Hyperspectral Plant Disease Assessments. IFAC Proceedings Volumes. 46:(4)361-365 http://www.sciencedirect.com/science/article/pii/S1474667016335728
Zack, G. W., W. E. Rogers, and S. A. LAtt. 1977. Automatic Measurement of Sister Chromatid Exchange Frequency. Histochemistry and Cytochemistry, 25(7), 741-753.
Zhang, M., Z. Qin, X. Liu, and S. L. Ustin. 2003. Detection of stress in tomatoes induced by late blight disease in California, USA, using hyperspectral remote sensing. International Journal of Applied Earth Observation and Geoinformation, 4(4), 295-310. doi: http://dx.doi.org/10.1016/S0303-2434(03)00008-4