(44.192.112.123) 您好!臺灣時間:2021/03/01 03:54
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:鄭琪鴻
研究生(外文):Chi-hung Cheng
論文名稱:一套鳳梨品種辨識系統之實現
論文名稱(外文):An Implementation of Pineapple BreedsRecognition System
指導教授:張鴻德張鴻德引用關係
學位類別:碩士
校院名稱:南台科技大學
系所名稱:資訊工程系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2010
畢業學年度:98
語文別:中文
論文頁數:72
中文關鍵詞:圖訊識別形態學處理紋理支持向量機
外文關鍵詞:Pattern RecognitionMorphologicalTextureSVM (Support Vector Machine)
相關次數:
  • 被引用被引用:0
  • 點閱點閱:237
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:1
近年來生物科技不斷的在進步,農作物的改良更是成為一大熱門,因此同樣
一種名稱的農作物,也會衍生出很多不同的種類,可借由外觀上的不同可以到達
辨認的目地,故本論文提出以紋理特徵及色彩的方式,實作於鳳梨品種辨識。本
論文有三個部份,一、影像的擷取,在訓練影像的部分,以鳳梨的果目作為訓練
影像,在測試的影像則是採用一整片的鳳梨表皮;二、特徵資料庫的建立,資料
庫是以五種台灣鳳梨各取一百個果目做為樣本將其色彩空間由RGB 轉換為
YCbCr,Y 值透過區域二元模式求得每個樣本的紋理特徵,並且加入Cb 與Cr 為
色彩特徵做為輔助,將得到的紋理與色彩特徵串聯成所代表之特徵向量,最後透
過分類器-支持向量機訓練出每種鳳梨果目之間的決策函數;三、特徵預測分類,
採用支持向量機依據特徵資料庫對於試影像中的所獲得的果目進行預測;其中,
果目切割的部分,首先對測試影像做對比增強技術增強果目與果實網狀的部分,
接著採用形態學將可能為果目影像位置一一的切割出來,採用與訓練樣本相同的
方法擷取其紋理和色彩特徵,根據先前所得到各種鳳梨果目之間的決策函數判定
此影像為何種鳳梨果目,最後統計所有切割出來的影像種類,以多數表決投票法
則來決定辨識結果。實驗證明本研究所提出之方法在鳳梨品種辨識上辨識率約為
80%。
In recent years, the biotechnology is keeping progressing, crop improvement especially.
Therefore, we can create many sorts of crops even thought they are in the same category.
We will identify their differences by different appearance of each. In this reason, this
thesis provides two methods of texture feature & color and distinguishes the difference
by experiment of pineapple. The thesis including three parts: First, “The Image
Capture”. In the image training part, I use pineapple eyes. In the image testing part, I
use a piece of pineapple coating; Second, “Establishment of Feature Database”. My
database is taking samples for 100 pineapple eyes from 5 kinds of Taiwan’s pineapples
and then transferring the color space from RGB to YCbCr. Y is equal to texture features
of each sample which got from local binary pattern. Besides, I add Cb and Cr as the
support of color feature. After that, I can link texture and color to what feature vector is.
Finally, getting decision function of each kind of quantity of pineapple’s eyes through
SVM.; Third, “Classification by Feature Prediction”. To use SVM according to the
feature database which established image testing on the pineapple eyes in order to
predict; In the above saying, the part of cutting pineapple’s eyes, I use image testing to
reinforce the part between pineapple eyes and net part by histogram stretching. After
that, I recognize all possible positions of pineapple eyes by morphological. Using the
same sample training method and getting texture & color feature. According to decision
function which got from previous image database of pineapple to judge what image is
for what kind of pineapple. Finally, gathering statistics from all images of segments and
deciding recognition result by vote. This experiment method was offered by my
research institute. It can prove that the recognition rate from pineapple breed is about
80%.
摘 要--------------------------------------------------------------------------------------------iv
Abstrasct --------------------------------------------------------------------------------------------- v
致 謝--------------------------------------------------------------------------------------------vi
目 次-------------------------------------------------------------------------------------------vii
圖目錄-----------------------------------------------------------------------------------------------ix
第一章 緒論-------------------------------------------------------------------------------------- 1
1.1 前言------------------------------------------------------------------------------------------ 1
1.2 研究動機與目的--------------------------------------------------------------------------- 1
1.3 相關文獻探討------------------------------------------------------------------------------ 4
1.4 論文架構------------------------------------------------------------------------------------ 5
第二章 系統流程與架構----------------------------------------------------------------------- 6
2.1 系統流程------------------------------------------------------------------------------------ 6
2.2 系統架構------------------------------------------------------------------------------------ 8
第三章 影像前處理---------------------------------------------------------------------------- 11
3.1 待測試影像--------------------------------------------------------------------------------12
3.2 灰階轉換-----------------------------------------------------------------------------------13
3.3 影像增強------------------------------------------------------------------------------------15
3.4 二值化--------------------------------------------------------------------------------------16
3.5 中值濾波-----------------------------------------------------------------------------------18
3.6 形態學處理--------------------------------------------------------------------------------19
3.6.1 侵蝕運算----------------------------------------------------------------------------------- 20
3.6.2 膨脹運算----------------------------------------------------------------------------------- 20
3.6.3 斷開運算----------------------------------------------------------------------------------- 21
3.6.4 區域填充----------------------------------------------------------------------------------- 21
3.7 連通區域標記-----------------------------------------------------------------------------23
3.8 找尋有效的鳳梨果目與影像切割-----------------------------------------------------25
viii
第四章 紋理及色彩特徵擷取---------------------------------------------------------------- 27
4.1 紋理特徵-----------------------------------------------------------------------------------27
4.1.1 區域二元模式---------------------------------------------------------------------------- 28
4.1.2 旋轉不變之區域二元模式------------------------------------------------------------ 30
4.1.2 LBP-HF-------------------------------------------------------------------------------------- 31
4.2 色彩特徵------------------------------------------------------------------------------------35
4.2.1 色彩空間----------------------------------------------------------------------------------- 35
4.2.2 RGB 色彩空間---------------------------------------------------------------------------- 36
4.2.3 HSI 色彩空間------------------------------------------------------------------------------ 37
4.2.4 YCbCr 色彩空間-------------------------------------------------------------------------- 39
4.3 鳳梨果目之紋理與色彩特徵比較-------------------------------------------------------- 40
第五章 分類辨識------------------------------------------------------------------------------- 44
5.1 支持向量機--------------------------------------------------------------------------------44
5.2 支持向量機之樣本訓練-----------------------------------------------------------------49
5.3 支持向量機之分類辨識------------------------------------------------------------------50
5.4 多數決投票法則---------------------------------------------------------------------------51
第六章 實驗結果與結論---------------------------------------------------------------------- 52
6.1 硬體環境-----------------------------------------------------------------------------------52
6.2 實驗一—正常光源下的辨識之辨識結果--------------------------------------------53
6.3 實驗二—外加光線干擾之辨識結果--------------------------------------------------56
6.4 實驗三—過熟的鳳梨之辨識結果-------------------------------------------------------- 57
6.5 實驗四—加入大量的背景結果--------------------------------------------------------59
6.6 實驗結論-----------------------------------------------------------------------------------60
參考文獻------------------------------------------------------------------------------------------- 61
1. Berbar, M.A.; Kelash, H.M.; Kandeel, A.A. ; ”Faces and Facial Feature Detection
in Color Images” Geometric Modeling and Imaging--New Trends, 2006.
2. 林家緯,以統計數據為基礎的車牌影像處理與辨識方法,國立中山大學機械
與機電工程學系研究所,民95 年。
3. 吳芳宇,指紋辨識系統,國立清華大學資訊工程學系碩士論文(2001)。
4. Ahonen T, Matas J, He C & Pietikäinen M. ; “Rotation Invariant Image
Description with Local Binary Pattern Histogram Fourier Features “ Image
Analysis, SCIA 2009 Proceedings, LNCS 5575, pp. 61-70.
5. 陳冠字,紋理分析在彩色影像分類上之應用,中華大學機械與航太工程研究
所碩士論文,民91年。
6. 陳幹旻,以特徵分佈為基礎之彩色紋理影像切割,元智大學資訊工程學系碩
士論文,民89年。
7. 程雅娟,結合色彩、紋理及區塊資訊之影像分類法,元智大學資訊工程學系
碩士論文,民90年。
8. 周哲帆,結合色彩與紋理特徵之人臉偵測及識別技術,清雲科技大學電子工
程研究所碩士論文,民97年。
9. Woo Chaw Seng; Mirisaee, S.H.; “A new method for fruits recognition system”
Electrical Engineering and Informatics, vol. 01, pp.130 - 134, 2009.
10. D. Unay; B. Goesslin ; “Automatic defect segmentation of 'Jonagold' apples on
multi-spectral images : A comparative study ” Postharvest biology and technology ,
vol. 42,Issue 3,pp.271-279,2006.
11. J. Serra, Image Analysis and Mathematical Morphology,Academic Press, London,
1982.
12. F. Chang C.J. Chen and C.J. Lu, “A Linear-Time Component-Labeling Algorithm
Using Contour Tracing Technique,” Computer Vision and Image Understanding,
vol. 93, no. 2, pp. 206-220, 2004.
13. Nielsen, B.; Albregtsen, F.; Danielsen, H.E. ; ”Low Dimensional Adaptive
Texture Feature VectorsFrom Class Distance and Class Difference Matrices”
Medical Imaging, IEEE Transactions on Volume 23, Issue 1,pp.73 – 84,2004.
14. Varma, M.; Zisserman, A., “A Statistical Approach to Texture Classification from
Single Images,” International Journal of Computer Vision, vol. 62, pp.
61-81.,2005.
15. M. Pietikäinen, T. Ojala, Z. Xu ; “Rotation-invariant Texture Classification Using
Feature Distributions,” Pattern Recognitiom 33,pp.43-52,2000
16. Zhang, J., Tan, T.: “Brief review of invariant texture analysis methods”. Pattern
Recognition , 2002 ,vol.35, pp.735–747.
17. Ojala, T., Pietikainen, M., Maenpaa, T.: “Multiresolution gray-scale and rotation
invariant texture classification with local binary patterns”. IEEE Transactions on
Pattern Analysis and Machine Intelligence , 2002 , vol 24 , pp.971–987.
18. Julius, T.T.; Rafael, C.G., “Pattern Recognition Principles,” Addison Wesley,
1974.
19. Jain, A.K.; Duin, R.P.W.; Jianchang Mao, “Statistical Pattern Recognition: A
Review,” Pattern Analysis and Machine Intelligence, 2000, vol 22, issue 1, pp.
4-37.
20. Vapnik, V.N., An overview of statistical learning theory. 1999. 10(5): p. 988.
21. Vapnik, V.N., Estimation of Dependencies Based on Empirical Data. 1982, NY:
Springer-Verlag.
22. http://www.csie.ntu.edu.tw/~cjlin/libsvm
連結至畢業學校之論文網頁點我開啟連結
註: 此連結為研究生畢業學校所提供,不一定有電子全文可供下載,若連結有誤,請點選上方之〝勘誤回報〞功能,我們會盡快修正,謝謝!
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔