臺灣博碩士論文加值系統

稻米是我國農產品中非常重要的一環，因此為了確保稻米的產量，病蟲害防治與治療更是重點問題。現今科技的發展之下，稻米高密度化是現今農業發展的主流，也造成了病蟲害發生時很難在早期發現並進行醫治，往往會導致大面積的感染使得稻米產量受到很大的影響。因此本論文希望透過影像處理的技術，設計出針對稻米病蟲害的辨識系統。
本論文首先對於稻米病蟲害的症狀進行歸納，由葉片病症作為出發點，分析出各項葉片病害的辨識關鍵性指標，最後再設計演算法進行辨識。對於測試的資料樣本的蒐集部分，因為現今尚未有龐大、可信的資料庫存在，因此本論文所蒐集的病害測試圖有限，因此總共十一種症狀表現形式中，本論文僅針對其中八種病害進行演算法開發，最後實際田野病害拍攝蒐集到五種共150張病蟲害測試圖進行驗證，最後所得到的辨識率為84.67%。

Rice is one of the most important agricultural products in ROC . In order to ensure the productivity of rice , the prevention and treatment of disease and pest is an important issue . Nowadays , the mainstream of rice cultivation is high density planting , witch make us hard to discover disease and pest of rice at the very first time . As a result , those plants with illness may infect other plants , and it will affect the total yield of rice . Therefore we design a recognition system based on image processing technology helping users recognize illness easily .

In our article , we first generalize the symptom of diseases and pest of rice . Then we choose leaf disease as the target , and analyze the key point of recognition of each illness with image . Therefore we may use some features to fit those points to recognize what illness occurred . In our research , one big difficulty is that there is no big and reliable database available from internet . It is hard for us to design a good algorithm from such a few samples . Overall , we propose a recognition system based on eight kind of symptoms , and we test our system with five kind of symptoms which include 150 pictures . Finally we get 84.67% matching rate for our testing .

摘要
Abstract
目錄
圖目錄
表目錄
第一章 緒論
1.1 研究動機
1.2 章節規劃
第二章 背景知識
2.1 稻米簡介
2.2 稻米病蟲害簡介
2.2.1 病害
2.2.2 蟲害
2.2.3 簡結
2.3症狀歸納
2.3.1 病斑
2.3.2 病紋
第三章 相關研究
3.1 針對特定病害辨識系統設計
3.2 顏色與形狀特徵
3.3 多層次式辨識系統
3.4 簡結
第四章 系統實現與演算法
4.1 數位影像與色彩空間
4.1.1 數位影像
4.1.2 RGB三原色光模式
4.1.3 YUV色彩空間
4.2 輸入影像格式
4.3 系統架構
4.3.1 系統整體架構
4.3.2 特徵參數萃取
4.3.3 病症資料庫與病症分析
4.3.4 中等距離多片葉分析
4.4 前置處理
4.4.1 Sobel 邊界演算法
4.4.2 侵蝕擴張法
4.4.3 色彩濾波-正常稻葉顏色
4.4.4 自動閥值二值化
4.4.5 分群演算法
4.5 特徵參數與葉緣檢測
4.5.1 群中心定義
4.5.2 形狀參數
4.5.3 顏色參數
4.5.4 葉尖檢測
4.5.5 葉緣檢測
4.6 病害辨識
4.6.1 病斑分析
4.6.2 病紋分析
4.6.3 K-means 分群法
4.6.4 特徵值檢測
第五章 實驗結果與數據
5.1 相機規格
5.2 辨識率分析
5.3 實際測試
5.4 相關成果比較
第六章 結論與未來展望
6.1 結論
6.2 未來展望
參考文獻

[1] 張淳淳，《水稻病蟲害圖鑑與管理手冊》，行政院農業委員會台南區農業改良場(2012)

[2]消巧玲、楊純明、李裕娟，〈水稻栽植密度對生長行為與穀粒產量之影響〉，行政院農業委員會農業試驗所(2009)

[3] Köhler, H.A , " Köhler's Medizinal-Pflanzen in naturgetreuen Abbildungen mit kurz erläuterndem Texte" , Gera-Untermhaus: Köhler, F.E.,(1887)

[4] 林慶元、施錫彬 等，《植物保護圖鑑系列8》，行政院農業委員會動植物防疫檢疫局(2007)

[5]YX Zhao, KR Wang, ZY Bai, SK Li, RZ Xie, SJ Gao, "Bayesian classifier method on maize leaf disease identifying based images " ,Comput. Engin. Applic., vol.43, no.5, pp.193-195, (2007)

[6]Santanu Phadikar & Jaya Sil,"Rice Disease Identification using Pattern Recognition Techniques", ICCIT 2008. 11th International Conference,(2008)

[7]S. Phadikar, J. Sil, & A. K. Das , "Classification of Rice Leaf Diseases Based on Morphological Changes", International Journal of Information and Electronics Engineering, Vol. 2, No. 3, (2012)

[8]Yingfeng Zhou, Yaming Wang, Qing Yao,"Segmentation of Rice Disease Spots Based on Improved BPNN", Image Analysis and Signal Processing (IASP), 2010 International Conference,(2010)

[9]Qing Yao, Zexin Guan, Yingfeng Zhou, Jian Tang, Yang Hu, Baojun Yang,"Application of support vector machine for detecting rice diseases using shape and color texture features", 2009 International Conference on Engineering Computation,(2009)

[10]Amrita A. Josh&B.D. Jadhav i," Monitoring and Controlling Rice Diseases Using Image Processing Techniques", Analytics and Security Trends (CAST), International Conference, (2016)

[11]Hemanta Kalita&Shikhar Kr. Sarma,"Expert System for Diagnosis of Diseases of Rice Plants: Prototype Design and Implementation", Automatic Control and Dynamic Optimization Techniques (ICACDOT), International Conference,(2016)

[12]MICHAEL W. SCHWARZ,"An Experimental Comparison of RGB, YIQ,LAB, HSV, and Opponent Color Models ", Transactions on Graphics (TOG) Volume 6 Issue 2, April 1987
Pages 123-158,(1987)

[13]D.H.Ballard, "Generalizing the Hough transform to detect arbitrary shapes", Pattern Recognition Volume 13, Issue 2, Pages 111-122,(1981)

[14]PartiEric Olson, "Particle Shape Factors and Their Use in Image Analysis", Journal of GXP Compliance; North Olmsted15.3: 85-96, (2011)

[15]Sreerupa Das1 & Michael Mozer, "Dynamic On-Line Clustering and State Extraction: An Approach to Symbolic Learning " Neural Networks Volume 11, Issue 1, Pages 53-64,(1998)

[16]聯合國稻米研究所international rice research institute (IRRI)

[17]廖紹綱，《數位影像處理》，東華書局股份有限公司，(2009)