跳到主要內容

臺灣博碩士論文加值系統

(44.212.99.208) 您好!臺灣時間:2024/04/17 18:25
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

: 
twitterline
研究生:姜岱昀
研究生(外文):TaiYun Chiang
論文名稱:基於兩段形狀特徵之影像品質驗證
論文名稱(外文):Shape-Based Image Quality Verification with Two-Stage Comparator
指導教授:陳金聖陳金聖引用關係
口試委員:黃正民李建德尹邦嚴
口試日期:2016-07-27
學位類別:碩士
校院名稱:國立臺北科技大學
系所名稱:自動化科技研究所
學門:工程學門
學類:機械工程學類
論文種類:學術論文
畢業學年度:104
語文別:英文
中文關鍵詞:條件機率拓樸網路區域特徵全域特徵形狀特徵影像品質驗證
外文關鍵詞:Conditional probabilityTree representation modelLocal featureGlobal featureShape featureImage quality verification
相關次數:
  • 被引用被引用:0
  • 點閱點閱:198
  • 評分評分:
  • 下載下載:5
  • 收藏至我的研究室書目清單書目收藏:0
印刷品質驗證技術廣泛應用於工業領域,大部分主要針對字元進行印刷品質驗證;現今提出之品質驗證方法對於多物件商標檢測效果不佳,因此本研究主要目的為提高多物件商標印刷品質驗證效率。影像品質驗證流程主要可分成前處理、影像切割、特徵擷取、比較器四大部分。本研究主要以特徵擷取與比較器為研究重點,探討全域與區域特徵擷取,並結合條件機率策略找到全域與區域特徵閥值,透過粗到細的比較策略進行影像品質驗證。
本研究方法在粗驗證階段時使用重心距離直方圖進行直方圖匹配,達到初步篩選之目的,當匹配率大於全域特徵閥值即進入細階段比較器;在細階段主要驗證物件的瑕疵、模糊以及物件與物件之間旋轉、位移關係。因此在區域特徵使用重心距離直方圖、區域重心距離直方圖、輪廓追隨直方圖與物件相關位移,以階層式的比較方式進行區域特徵驗證,實驗結果顯示本論文所提出的方法可有效地驗證影像品質。
Printed image quality verification is often applied in the industrial applications and mainly used to verify the optical character. The methods which have been proposed are inefficient for the trademark; therefore, the purpose of this thesis is to improve the verification efficiency for the trademark images. The proposed verification technique uses the coarse-to-fine strategy, which verifies the quality of composite images using global feature and local features.
The proposed algorithm contains a training phase and a verification phase. In the training phase, the similarity thresholds for the global feature and local features are automatically calculated by the conditional probability algorithm. The global feature are described using a histogram of centroid distances (HCD). The local features, including the HCD, the regional histogram of the centroid distance (RHCD), the contour-shifted profile and the object relative shift, are described using a tree presentation model (TRM). In the verification phase, a coarse-to-fine strategy is employed to speed up the verification process. The global similarity is calculated at the course stage, using the HCD for the entire image. When the global similarity is greater than the threshold, the candidate objects are passed into the fine stage. A hierarchical evaluation is then used to compare the topological similarity at the fine stage. The experimental results show that the proposed method accurately verifies the quality of images.
摘 要 i
ABSTRACT ii
致謝 iii
Contents iv
List of Tables vi
List of Figures vii
Chapter 1 Introduction 1
1.1 Motivation 1
1.2 Previous work 2
1.3 Research Method 3
1.4 Organization of this thesis 4
Chapter 2 Feature Extraction 6
2.1 Global feature extraction 8
2.2 Tree representation model 10
2.3 Local feature extraction 14
2.3.1 Histogram of centroid distance 14
2.3.2 Regional histogram of the centroid distance 15
2.3.3 Contour-shifted profile 16
2.3.4 Object relative shift: 18
Chapter 3 Conditional Probability Algorithm 19
Chapter 4 Coarse-to-Fine Comparator 23
4.1 Coarse Stage Comparator 25
4.2 Fine Stage Comparator 25
4.2.1 Tree representation model structure comparison 25
4.2.2 Histogram of centroid distance comparison 27
4.2.3 Regional histogram of the centroid distance comparison 27
4.2.4 Contour-shifted profile comparison 28
4.2.5 Object relative shift comparison 30
Chapter 5 Experimental Results 32
5.1 Rotation-invariant experiment 32
5.2 Experimental parameters set up 35
5.3 Evaluation methodology 40
5.4 Image quality verification experiment 40
5.4.1 Case 1 41
5.4.2 Case 2 46
Chapter 6 Conclusion 49
Reference 50
Nomenclature 53
[1] Elias N Malamas, Euripides G.M Petrakis, Michalis Zervakis, Laurent Petit and Jean-Didier Legat. , “A survey on industrial vision systems”, Image and vision computing, vol. 21, no. 2, 2003, pp. 171-88.
[2] Yi Lu, “Machine Printed Character Segmentation - An Overview”, Pattern Recognition, vol. 28, no. 1, 1995, pp. 67-80.
[3] Bernard C. Jiang, Szu-Lang Tasi and Chien-Chih Wang, “Machine vision-based gray relational theory applied to IC marking inspection”, IEEE Transactions on Semiconductor Manufacturing, vol. 15, no. 4, 2002, pp. 531-539.
[4] R. Nagarajan, Sazali Yaacob, Paulraj Pandian and Marzuki Karthogayan, “A real time marking inspection scheme for semiconductor industries”, The International Journal of Advanced Manufacturing Technology, vol. 34, no. 9, 2006, pp. 926-932.
[5] Open eVision Eval (Version 1.2), [Euresys Inc.]. Available online: http://www.euresys.com/product/easyocv/?tabproduct=benefits (accessed on 3 June 2016)
[6] Shin-Chieh Lin and Chun-Lin Yu. , “Analysis of IC Image Alignment and Marking Defect Inspection”, Master’s thesis, National Tsing Hua University, Hsinchu, Taiwan, 2006.
[7] Sung-Hyuk Cha and Sargur N. Srihari., “Multiple feature integration for writer verification”, Proceedings of the 7th International Workshop on Frontiers in Handwriting Recognition, Sep. 2000, pp. 333-342.
[8] Narit Hnoohom, Narumol Chumuang and Mahasak Ketcham, “Thai Handwritten Verification System on Documents for the Investigation”, 11th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS), 2015, pp. 617-622.
[9] Heng Qi, Keqiu Li, Yanming Shen and Wenyu Qu, “An effective solution for trademark image retrieval by combining shape description and feature matching”, Pattern Recognition, vol. 43, no. 6, 30 Jun 2010, pp. 2017-2027.
[10] Rafael C. Gonzalez and Richard E. Woods, Digital Image Processing, 3rd edition, Prentiz-Hall, Inc.: Upper Saddle River, NJ, 2006.
[11] Philippe Salembier and Luis Garrido., “Binary Partition Tree as an Efficient Representation for Image Processing Segmentation and Information Retrieval”, Image Processing, vol. 9, no. 4, Apr 2000, pp. 561-576.
[12] Pascal Monasse and Frédéric Guichard, “Fast computation of a contrast-invariant image representation”, IEEE Transactions on Image Processing, vol. 9, no. 5, May 2000, pp. 860-872.
[13] Naif Alajlan, Mohamed S. Kamel and George H. Freeman, “Geometry-based image retrieval in binary image databases”, Pattern Analysis and Machine Intelligence, vol. 30, no. 6, Jun 2008, pp. 1003-1013.
[14] Dong Liu, Shengsheng Wang, Yiting Liu, Fantao Zeng, Jimin Wu and Wenyang Li, “Tree Representation and Feature Fusion Based Method for Multi-object Binary Image Retrieval”, Journal of Information and Computational Science, vol. 10, no 4, 2013, pp. 1055-1064.
[15] Miha Možina , Dejan Tomaževi , Franjo Pernuš and Boštjan Likar, “Optical character verification of print on pharmaceutical capsules”, 12th IAPR Conference on Machine Vision Applications, 2011, pp. 303-306
[16] Antonio Rama, Ramon Alujas and Francesc Tarres, “Fast and Robust Graphic Character Verification System for TV Sets”, Eighth International Workshop on Image Analysis for Multimedia Interactive Services, June 2007, pp.6-8.
[17] John Canny, “A computational approach to edge detection”, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 8, no.6, 1986, pp. 679-98.
[18] Pitrelli J. Fpitrelli and Michael Peter Perrone, “Confidence modeling for verification post-processing for handwriting recognition”, Proceedings of the 8th International Workshop on Frontiers in Handwriting Recognition, 2002, pp. 30-35.
[19] Henry Scheffé, “The analysis of variance”, Vol. 72, John Wiley & Sons, 1999.
[20] Roger W. Hoerl, “Six Sigma and the future of the quality profession”, Quality progress, vol. 31, no. 6, Jun 1998, pp.35-39.
[21] Breyfogle III, ForrestW, Implementing six sigma: smarter solutions using statistical methods, 2nd edition, John Wiley & Sons, 2003.
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