跳到主要內容

臺灣博碩士論文加值系統

(18.97.14.82) 您好!臺灣時間:2025/02/07 03:49
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:林姝吟
研究生(外文):Shu-Yin Lin
論文名稱:模糊類神經在二維物體辨識之應用
論文名稱(外文):Applying Fuzzy Neural Network Theory to Two-dimensional Objects Recognition
指導教授:吳 文 言
指導教授(外文):Wen-Yen Wu
學位類別:碩士
校院名稱:義守大學
系所名稱:管理科學研究所
學門:商業及管理學門
學類:企業管理學類
論文種類:學術論文
論文出版年:2000
畢業學年度:88
語文別:中文
論文頁數:62
中文關鍵詞:機器視覺支配點偵測特徵分類模糊K鄰近集群分析法倒傳遞類神經網路物體辨識重疊物體辨識
外文關鍵詞:machine visiondominant point detectionfeature classificationfuzzy k-nearest-neighborback-propagation neural networkobject recognitionoccluded object recognition
相關次數:
  • 被引用被引用:1
  • 點閱點閱:1125
  • 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:1
近幾年來,機器視覺(machine vision)應用在物體影像自動辨識與自動化檢驗的系統中,有了迅速的發展,使用的範圍也越來越廣闊,本研究著重在於待測點根據每點不同的k值找出適當的支配點位置和數量之間的關係。
除此之外,我們還提出一套結合模糊K鄰近集群分析法(fuzzy K-nearest-neighbor)和倒傳遞類神經網路(back-propagation neural network )的物體影像辨識系統。主要分成下列幾個步驟,首先取得物體的影像後,偵測輪廓支配點(dominant point),並計算連續兩個支配點與多邊形的重心所形成的三角形的邊長、緊密度及連續三個支配點所形成三角形的夾角餘弦值、緊密度(compactness)當作特徵值。接著運用模糊k鄰近集群分析法將特徵值作適當的分類,再將每類的所有特徵值計算其重心點以為該類的代表性特徵值,如此可以減少測試物的特徵值數目,也可以更方便快速的辨識出物體,節省很多辨識的時間。最後將分類後的特徵值代入倒傳遞類神經網路模式中訓練測試,以辨識出正確的影像。本研究以兩組不同影像進行辨識,由實驗結果得知,此方法可以有效地辨識物體,第一組影像的辨識時間約為5.38秒,第二組影像的辨識時間約為3.3秒,總平均辨識率約為95%。
同樣地本研究運用本文提出辨識方法辨識重疊物體影像,主要的方法首先偵測重疊物體影像的支配點,並計算兩個支配點與多邊形的重心所形成的三角形的邊長,以及計算連續三個支配點所形成三角形的夾角餘弦值當作特徵值。再運用倒傳遞類神經網路將單一物體影像所有特徵值輸入網路學習訓練之後,把重疊物體影像的特徵值輸入進行比對分類,最後將一重疊物體影像的特徵值分為一到多個新的未知物體影像,再次利用倒傳遞類神經網路將單一物體影像特徵值輸入類神經網路訓練學習後,把這些未知影像的特徵值輸入網路測試,最後可以辨識出來重疊物體影像中包含哪些單一物體影像。實驗的結果得知本研究的辨識方法可以將一重疊物體影像分離為一到數個單一物體影像,並得知每一個單一物體影像在一重疊物體影像可能的機率值。

Recently, machine vision has been applied in recognition and inspection. In this paper we proposed the method to detect the dominant points which do not input the fix value of the region of the forward and backward points in advance. This method can detect the dominant points of the two dimension objects efficiently and automatically.
Besides, we also proposed a method based on the fuzzy K-nearest neighbors clustering and back-propagation neural network method to recognize the two-dimensional objects. There are four steps of the proposed method. First, detect the dominant points of the images. Secondly, we calculate the length and the compactness of the triangle of each two continuous dominant points and the center of the polygonal object ,and the cosines value and compactness of the triangle of each three continuous dominant points as the feature of the object. Thirdly, use the fuzzy K-nearest neighbors clustering to classify the feature values. Finally, the back-propagation neural network model has been adopted to recognize the objects. The experimental results indicate that the proposed method has about 95% of correct recognition rates and the recognition time are 5.38 second in first part images and 3.3 second in second part images.
We use the method to recognize the occluded objects. First, we detect the dominant points of the occluded objects. Secondly, we calculate the length of the triangle of each two continuous dominant points and the center of the polygonal object ,and the cosines value of the triangle of each three continuous dominant points as the feature of the object. Thirdly, we input the features of the simple objects as the train data into the back-propagation neural network and class the features of the occluded objects. Fourthly, after classed these features, we select the feature of the same class as the features of a new simple object. Then we input the features of the simple objects as the train data into the back-propagation neural network and the features of the new simple objects as the test data. Finally, we can know that the occluded object which includes by what simple objects and get the probability of the simple object cover the occluded objects.

中文摘要 Ⅰ
英文摘要 Ⅱ
目錄 Ⅲ
表次 Ⅴ
圖次 Ⅵ
第一章 緒論 1
第一節 研究動機 2
第二節 研究目的 2
第三節 研究方法概述 3
第三節 研究方法流程 4
第二章 文獻探討 6
第一節 支配點偵測 6
第二節 特徵值擷取 8
第三節 特徵值分類 9
第四節 物體辨識 9
第五節 類神經網路簡介 11
第三章 影像前處理 15
第一節 影像前處理方法 15
第二節 適應性支配點偵測 17
第四章 特徵值擷取與分類 25
第一節 物體值擷取 25
第二節 物體值分類26
第五章 物體辨識 29
第一節 倒傳遞類神經網路30
第二節 單一物體影像辨識流程 34
第三節 重疊物體影像辨識流程 35
第六章 實驗結果與分析 37
第一節 實驗設備簡介37
第二節 支配點偵測38
第三節 單一物體影像辨識 41
第四節 重疊物體影像辨識 47
第七章 結論與未來展望52
參考文獻 53

1. 吳嘉賓(1995),應用時空型類神經網路於部分遮蓋物之辨識,碩士 論文,成功大學電機工程研究所。
2. 溫永豐(1992),以輪廓特徵做平面圖形辨識,碩士論文,成功大學機械工程研究所。
3. 葉怡成(1998),類神經網路模式應用與實作,儒林出版社。
4. 廖博仁(1994),部份重疊工件辨識:輪廓點特徵比對,碩士論文,元智工學院工業工程研究所。
5. 廖嘉楷(1994),部份重疊工件辨識:幾何型態比對,碩士論文,元智工學院工業工程研究所。
6. 蔡瑞源(1995),以類神經網路排定部份重疊工件辨識之順序,碩士論文,元智工學院工業工程研究所。
7. 鄧博文(1998),模糊集群神經網路模式之建立與應用,碩士論文,義守大學管理科學研究所。
8. Abe, V. S. and M. S. Lan(1995), “ A nearest-network-based fuzzy classifier”, IEEE Trans. on Systems man and cybernetics, 25, pp.353-361.9. Chen, C. L. and R. S. Nutter(1991), “Improving the training speed of three-layer feedforward neural nets by optimal estimation of the initial weight”, Neural Networks, 3, pp.2063-2068.
10. Chen, C. L. and R. S. Nutter(1992), “An extended back-propagation learning algorithm by using heterogeneous processing units”, Neural Networks, 3, pp.988-993.
11. Chung, P. C., E. L. Chen, and C. T. Tsai(1994), “ Pattern recognition using a hierarchical neural network”, IEEE International Conference on Neural Networks, 5, pp.3104-3109.
12. Chung, P. C., E. L. Chen, and J. B. Wu(1998), “ A spatiotemporal neural network for recognizing partially occluded objects”, IEEE Trans. on Signal Processing, 46, pp.1991-2000.
13. Cornic, P.(1997), “Another look at the dominant point detection of digital curves”, Pattern Recognition Letters, 18, pp.13-25.
14. Freeman, H. and L. S. Davis(1977), “A corner-finding algorithm for chain-code curves”, IEEE Transactions on Computers, 26, pp.297-303.
15. Gates, G. W.(1972), “ The reduced nearest neighbor rule”, IEEE Trans. on Information Theory, 3, pp.431-433.
16. Gonzalez, R. C. and R. E. Wood(1993), Digital Image Processing, Addison-Wesley, USA.
17. Han, M. H. and D. Jang(1990), “Use of maximum curvature points for the recognition of partially occluded objects”, Pattern Recognition, 23, pp.21-33.
18. Hattori, K. and M. Takahashi(1999), “A new nearest-neighbor rule in the pattern classification problem”, Pattern Recognition, 32, pp.425-432.
19. Hsieh, K. R., and W. T. Chen(1993), “A neural network model which combines unsupervised and supervised learning”, IEEE Transactions on Neural Networks, 4, pp.357-360.
20. IP-LIB Reference Manual, Matrox Electronic System Ltd., 1991.
21. Keller, J. M., Gray, M. R. and J. G. JR(1985), “A fuzzy k-nearest neighbor algorithm”, IEEE Trans. on System, Man, and Cybernetics, 4, pp.580-585.
22. Kim, D. S., and T. L. Huntsberger(1991), “Self-organizing neural networks for unsupervised pattern recognition”, Tenth Annual International Phoenix Conference on Computers and Communications, pp.39-45.
23. Kuo, C. M., C. H. Hsieh, and C. P. Chao(1995), “Image segmentation based on a new fuzzy c-means clustering approach”, Academic Bulletin of Chinese Naval Academy, 5, pp.29-36.
24. Lai, J. Z.C. and J. M. Lin(1991), “Recognizing partially occluded 2-D parts based on the tracing of feature points”, in Proceedings - IEEE International Conference on Robotics and Automation, Piscataway, NJ, USA, pp.1796-1801.
25. Liu, H. C. and M. D. Srinath(1990), “Partial shape classification using contour matching in distance transformation”, IEEE Trans. Pattern Analysis and Machine Intelligence, 12, pp.1072-1079.
26. Nirwan A. and E. J. Delp(1991), On detecting dominant points, Pattern Recognition, 24, 5, 441-451.
27. Ogawa, H.(1994), “Fuzzy relaxation technique for partial shape matching”, Pattern Recognition Letters, 15,4, pp.349-355.
28. Ramer, U.(1972), “An iterative procedure for the polygonal approximation of plane curves”. Computer Graphics Image Process. 1, pp.244-256.
29. Ray, B. K. and K. S. Ray(1992), “An algorithm for detection of dominant points and polygonal approximation of digitized curves”, Pattern Recognition Letters, 13, pp.849-856.
30. Ray, B. K. and K. S. Ray(1992), “Detection of significant points and polygonal approximation of digitized curves”, Pattern Recognition Letters, 13, pp.443-452.
31. Rosefeld, A. and E. Johnston(1973), “Angle detection on digital curves”, IEEE Transactions Computers, 22, pp.875-878.
32. Rosenfeld, A. and I.S. Weszka(1975), “An improved method of angle detection of digital curve”, IEEE Transactions on Computers, 24, pp.940-941
33. Ross, T. J.(1995), Fuzzy Logic with Engineering Applications, McGraw-Hill.
34. Sanchiz, J. M., F. Pla and J. M. Inesta(1998), Using neural network to detect dominant points in chain code contours, Pattern Recognition and Artificial Intelligence, 12, 5, pp. 661-675.
35. Sankar, P. V. and C. V. Sharma(1978), “A parallel procedure for the detection of dominant points on digital curve”, Computer Graphics and Image Processing, 7, pp.403-412.
36. Sherif, M. S., and M. S. Samee(1994), “Pattern recognition using neural networks that learn from fuzzy rules”, Proceeding of the 37th Midwest Symposium, 1, pp.599-602.
37. Sheu, H. T. and W. C. Hu(1996), A rotationally invariant two-phase scheme for corner detection, Pattern Recognition, 29, 5 , 819-828.
38. Shi, Y. H., and Z. Y. He(1991), “New neural network for pattern recognition”, Electronics Letters, 27, pp.1353-1355.
39. Ten, Cho-Huak, and Poland T. Chin(1989), “On the detection of dominant points on digital curves”, IEEE Transactions on Analysis and Machine Intelligence, 11, pp.859-872.
40. Tsang, P. W.M. and P.C. Yuen(1993), “Recognition of partially occluded objects”, IEEE Trans Systems, Man and Cybernetics, 23, pp.228-236.
41. Turney J. L., T. N. Mudge and R. A. Volz(1985), “Recognizing partially occluded parts”, IEEE Trans. on Pattern Analysis and Machine Intelligence, vol PAMI-7, pp.410-421.
42. Wang, D.(1993), “Pattern recognition: neural networks in perspective”, IEEE Expert, 8, pp.52-60.
43. Wang, M. J., W. Y. Wu(1995), L.K. Huang and D. M. Wang, “Corner detection using bending value”, Pattern Recognition Letters, 16, pp.575-583.
44. Wu, W. Y., and M. J. Wang(1993), “Detecting the dominant points by the curvature-based polygonal approximation”, CVGIP Graphical Model and Image Processing, 55, pp.79-88.
45. Wu, W. Y., and M. J. Wang(1999), “Two-dimension object recognition through two-stage matching”, IEEE Trans. on Image Processing, 8, pp.978-981.
46. Zahid, N., O. Abouelala, and M. A. Essaid(1999), “Unsupervised fuzzy clustering”, Pattern Recognition Letters, 20, pp.123-129.
47. Zheng, Nanning, Yaoyong Li, and Wiek P. M. Houwers(1995), “Local feature-based recognition of partially occluded objects using neural network”, in IECON Proceedings (Industrial Electronics Conference) v 2 1995. IEEE, Los Alamitos, CA, USA, pp.1301-1306.

QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊