跳到主要內容

臺灣博碩士論文加值系統

(3.81.172.77) 您好!臺灣時間:2022/01/21 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

我願授權國圖
: 
twitterline
研究生:張倍魁
研究生(外文):Chang Bei-Kuei
論文名稱:名片商標之擷取與辨識
論文名稱(外文):Extraction and Recognition of Logos in Name Cards
指導教授:李錫堅李錫堅引用關係
指導教授(外文):Lee, His-Jian
學位類別:碩士
校院名稱:國立交通大學
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:2002
畢業學年度:90
語文別:英文
論文頁數:58
中文關鍵詞:辨識
外文關鍵詞:recognitionlogo
相關次數:
  • 被引用被引用:1
  • 點閱點閱:282
  • 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
在日常生活中名片被廣泛的用在保存個人資訊. 名片中的商標則是具有獨特的象徵意義.在這篇論文中, 我們設計了一個名片商標自動擷取與辨識的系統.因此我們可以辨別出名片之中的商標並把商標儲存到資料庫內. 我們系統的輸入是二值化過後沒有雜訊的影像. 系統包含兩個主要的模組:商標的擷取和商標的辨識. 再商標擷取的模組中, 我們首先使用了更快且更適合名片影像的改良式Run-Length Smoothing Algorithm (RLSA)已減少商標候選個數. 接著, 我們提出可能影響商標候選的五個調整: 對符合姓名欄位的商標候選減少商標可能性. 對符合文字行特性的商標候選減少商標可能性. 對含有較少連接單元之商標候選增加其商標可能性.對於含有大量連接單元的商標候選減少其商標可能性.對於面積小的商標候選減少其商標可能性. 最後我們選有最高商標可能性的商標候選為我們的商標.在商標辨識模組中, 我們設計了一個對名片商標具有高辨識率的辨識引擎. 我們用了三個個特徵值: contour directional features, crossing count features and peripheral background area features. 在我們的商標擷取實驗中, 我們選取了610張具有商標的名片. 成功率為82.9%. 在我們的商標辨識實驗中, 我們選取了187個公司共495個商標. 成功率為 96.9%

Name cards are used widely for keeping personal information in our daily life. Logos in name cards are designed to represent the corporation recognition mark. In this thesis, we design an automatic extraction and recognition tool for logos in name cards. Thus, we can identify logos in name cards and store logos in a database. The input of our system is binary name card image without noise. Our system contains two major modules: logo extraction and logo recognition. In the logo extraction module, we first use an improved Run-Length Smoothing Algorithm (RLSA), which is faster and more suitable for name card images to reduce candidates of logos. Next, we adjust the logo possibility as follows: Reduce the logo possibility of the smeared CCs if they satisfy the name field constraint, Reduce the logo possibility of the smeared CCs which are collinear in vertical or horizontal direction and have similar width or width in the orthogonal direction, Increase the logo possibility of a smeared CC if its size is similar to the size of the connected components in the original (un-smeared ) image, Reduce the logo possibility of smeared CCs which contain many small connected component in the original (un-smeared) image and Reduce the logo possibility of smeared CCs which are very small to calculate the score of the candidates. Finally, we choose the highest score of the candidates of the logo as a logo. In the logo recognition module, we design a good recognition engine which has a high recognition rate for logos in the name card. We use three features: contour directional features, crossing count features and peripheral background area features. In our logo extraction experiments, we select 610 images of name cards. The success rates are 82.9% in logo extraction. In our logo recognition experiments, we select 495 logo images of 187 companies. The success rates are 96.9% in logo recognition.

TABLE OF CONTENTS
ABSTRACT (IN CHINESE)…………………………………………....i
ABSTRACT (IN ENGLISH)…………………………………………..ii
TABLE OF CONTENTS…………………………….………………...iv
LIST OF FIGURES……………………………………………………vi
LIST OF TABLES…………………………………………………….viii
CHAPTER 1. INTRODUCTION 1。
1.1 Motivation 1。
1.2 Survey of related Research 2。
1.3 Statistics of logos 3。
1.4 System Description and Assumption 5。
1.4.1 System Description 6。
1.4.2 Assumption 7。
1.5Thesis Organization 7。
CHAPTER 2. PREPROCESSING 9。
2.1 Block Segmentation 9。
2.1.1 Top-down Method 9。
2.1.2 Bottom-up Method 10。
2.1.3 Run-length Smearing Algorithm (RLSA) 11。
2.2 Proposed Method 12。
2.3 Non-logo component removal 15。
CHAPTER 3. LOGO EVALUATION 17。
Adjustment 1: Reduce the logo possibility of the smeared CCs if they satisfy the name field constraint 18。
Adjustment 2: Reduce the logo possibility of the smeared CCs which are collinear in vertical or horizontal direction and have similar width or width in the orthogonal direction 21。
Adjustment 3: Increase the logo possibility of a smeared CC if its size is similar to the size of the connected components in the original ( un-smeared ) image 23。
Adjustment 4: Reduce the logo possibility of smeared CCs which contain many small connected component in the original (un-smeared) image 27。
Adjustment 5: Reduce the logo possibility of smeared CCs which are very small 29。
Example: 31。
CHAPTER 4: LOGO RECOGNITION 36。
4.1 Motivation 36。
4.2 Evaluation of Statistical Features 36。
4.2.1 Uniform segmentation of logos 37。
4.2.2 Contour direction features(CD)…………………………………………... 38
4.2.3 Crossing counts features (CC) 39。
4.2.4 Peripheral background area (PBA) 40。
4.3 Measurements of Feature Distance 41。
4.4 Feature weight 42。
CHAPTER 5: EXPERIMENT RESULTS AND ANALYSIS 44。
5.1 Introduction 44。
5.2 Experiments of logo evaluation 44。
5.2.1 Detection of name cards with a logo and without a logo 44。
5.2.2 Correctness of logo evaluation 45。
5.2.3 Analysis for logo evaluation 49。
5.3 Experiment of logo recognition 52。
5.3.1 Correctness of logo recognition 52。
5.3.2 Warning rate of logos not in our database 53。
5.3.3 Analysis of logo recognition 53。
CPAPTER 6: CONCLUSION AND FUTURE WORK 55。
References 56。

References
[1] Ha Jaekyu. R. M. Haralick and I. T. Phillips, “Recurisive X-Y cut using bounding boxes of connected components,” Proc. Of the Int. Conf. On Document Analysis and Recognition, Vol. 2, pp. 952-955, Aug., 1995.
[2] Shih F. Y., S. S. Chen, D. C. D. Hung and P. A. Ng, “A document segmentation, classification system,” Proc. Of 2nd Int. Conf. Systems Integration, pp. 258-267, June, 1992.
[3] Y Lu, “Machine Printed Character Segmentation- an Overview,” Pattern Recognition, vol. 28, no. 1, pp. 67-80, 1996.
[4] Y.Lu, and M. Shridhar, “Character Segmentation in Handwritten Words — an Overview,” Pattern recognition, vol. 29, no. 1, pp. 77-96, 1996.
[5] Rechard G. Casey and Eric Lecolinet, “A survey of Methods and Strategies in Chinese Segmentation,” IEEE Trans. Pattern Anal. Mach. Intell.,vol.18, no. 7, pp.690-706, 1996
[6] Frank Y.Shih and Shy-Shyan Chen, “Adaptive Document Block Segmentation and Classifucation,” IEEE Trans. On System, Men, And Cybernetics. Vol. 26, no. 5. 1996
[7] L. O’Gorman, “The document spectrum for page layout analysis,” IEEE Trans. Pattern Anal. Machine Intell. Vol 15, pp.1162-1173, 1993.
[8] F. Wahl, K. Wong, R. Casey, “Block segmentation and text evaluation in mixed text/image document,” Graphics Modules Image Process. 20(1982) 375-390.
[9] K. Kise, A. Sato, M. Iwata, “Segmentation of page images using the area Voronoi diagram,” Comput. Vision Image Understanding. 70(1998) 370-380.
[10] A. Fletcher, R. Kasturi, “A robust algorithm for text string separation from mixed text/graphics images,” IEEE Trans. Pattern Anal. Machine Intell. 10(1998) 910-918.
[11] G. Nagy, S. Seth, M. Viswanathan, “A prototype document image analysis system for techical journal,” Computer 25(1992) 10-22.
[12] H. S. Baired, “Background structure in document image,” Int. J. Pattern Recognition Artif. Intel. 8 (1994) 1013-1030.
[13] H. S. Baired, S. E. Jones, S. J. Fortune, “Image segmentation by shape-directed covers,” International conference on Pattern Recognition. 1990, 820-825.
[14] S. L. Zhao, “Multi-kernal Chinese Character Recognition and A Simplified Language Module Used in Gerenal Document Processing system,” Master thesis, Institute of Computer Science and Imformation Enigineering, National Chiao Tung University, Taiwan,R.O.C., 2000.
[15] F. H. Cheng and W. H. Hsu, “Research on Chinese OCR in Taiwan,” Internation journal of Pattern Recognition and Artificial Intelligence, vol. 5 nos. 1ž, pp.139-164, June 1991
[16] I. C. Jou, Y. G. Chau, R. H. Ju, and J. Y. Lee,”Moduling for Chinese character pattern,” Proc. CVGIP, Taiwan, R.O.C., 1994, PP. 131-134.
[17] Bezdek, J. C., “Pattern Recognition with Fuzzy Objective Fiction Algorithm,” Plenum Press, New York 1981.
[19] Tou, G., Gonzalez, R., “Pattern Recognition Principle,” Addition-Wesley, Reading, MA. 1987.
[20] Nikhil R. Pal and Snakar K. Pal, “A Preview on Image Segmentation Technique,” Pattern Recognition, Vol. 26, NO. 9, pp. 1277-1294, 1993.
[21] Sun W., Lia L. M., et al. “Intelligent OCR processing,” journal of the American Society for Imformation Science, Vol. 43, No. 6, pp. 422-431, 1992.
[22] D. Wang, S. N. Srihari, “Classification of newspaper image block using texture analysis,” Compute. Vision Graphics Image Process. 20 (1989) 327-352.
[23] Y. H. Chiou, “Recognition of Chinese Business,” Master thesis, Institute of Computer Science and Imformation Enigineering, National Chiao Tung University, Taiwan,R.O.C., 1996.
[24] S. Mori, K. Yamamoto, “Research on machine recognition of handprinted characters,” IEEE Transations on Pattern Analysis and Machine Intelligence, vol. PAMI-6, no. 4, pp. 386-405. 1984.
[25] Y. H. Tseng, K. C. Kuo and H. J. Lee, “Speeding up Chinese Character Recognition in an Automatic Document Reading System,” Proc. Fourth Int. Conf. Document Analysis and Pecognition, pp.629-632, Ulm. Germany, 1997.

QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top