跳到主要內容

臺灣博碩士論文加值系統

(54.224.133.198) 您好!臺灣時間:2022/01/27 05:27
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:吳杭芫
研究生(外文):Hung-Yuan Wu
論文名稱:利用連接元件分析來做文件拼圖
論文名稱(外文):Document Mosaic via Connected-Component Analysis
指導教授:李錫堅李錫堅引用關係
指導教授(外文):Hsi-Jian Lee
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:中文
論文頁數:73
中文關鍵詞:連接元件文件拼圖
外文關鍵詞:connected-componentdocument mosaic
相關次數:
  • 被引用被引用:0
  • 點閱點閱:222
  • 評分評分:
  • 下載下載:29
  • 收藏至我的研究室書目清單書目收藏:1
在本篇論文中,我們建立一套文件合併系統用以解決在一般中文報紙合併時所遇到的一些問題。這些問題包括了文字的合併、標題的合併、表格的合併、圖形的合併等。
首先,所要處理的文件可能會傾斜,這種傾斜可能會影響到切字及辨識,我們嘗試使用一種方法偵測傾斜角度並提出一個快速的傾斜校正方法來旋轉文件影像。在這個系統中,為了做各種形式合併,我們必須先實行文件分析來認知文字資訊、標題資訊、表格資訊、圖形資訊。
在文件合併方面,我們有兩個方法,第一個方法是:我們利用在文章裡每個未辨識中文字的連通元件(Connected-Components)數目不同的特性來找出文件相同處,進而利用此相同處合併它們。第二個方法是:我們利用在文章裡每個已辨識中文字的唯一獨特性來找出文件相同處,進而利用此相同處合併它們。
在標題合併方面,我們也是利用在標題中每個未辨識中文字的連通元件(Connected-Components)數目不同的特性來找出文件相同處,進而利用此相同處合併它們。
在表格合併方面,我們首先刪除表格裡的水平和垂直線,於是我們得到了文字部分,所以利用在文章裡每個未辨識中文字的連通元件(Connected-Components)數目不同的特性來找出文件相同處,進而利用此相同處合併它們。
在圖形合併方面,我們抽取在圖形內一些較大連通元件(Connected-Components),然後比較這些較大連通元件(Connected-Components)的長度與寬度之大小來找出圖形相同處,進而利用此相同處合併它們。
以我們的實驗結果來說,這套文件合併系統有不錯的表現。

In this thesis, we construct a document matching and merging system to solve several problems in dealing with Chinese newspapers. These problems include matching and merging of plain texts, matching and merging of titles, matching and merging of tables, matching and merging of pictures, and matching and merging of mixed type images.
Input images probably have skew angles. These skew angles will affect the performance of character segmentation and character recognition. A skew angle detection method is proposed and a rotation transform is used to rotate document images. For the matching and merging of Chinese newspapers, we first perform document layout analysis to get text block, title block, table block, and picture block.
For the matching and merging of plain texts, we propose two methods. The first method is about the feature of the number of connected-components in a character. In most cases, different characters have different values of the number of connected-components in characters. We utilize the property to find matching between overlapped text images and merge them as a bigger text image. The second method is about the feature that a Chinese character is a basic unit. It is very simple to compare two Chinese characters. We utilize the property to find matching between overlapped texts and merge them as a bigger text.
For the matching and merging of titles, our method is about the feature of the number of connected-components in a character. In most cases, different characters have different values of the number of connected-components in characters. We utilize the property to find matching between overlapped title images and merge them as a bigger title image.
For the matching and merging of tables, we first erase horizontal and vertical lines in the table and translate the table into the plain text image. And we use a method about the feature of the number of connected-components in a character. In most cases, different characters have different values of the number of connected-components in characters. We utilize the property to find matching between overlapped tables and merge them as a bigger table.
For the matching and merging of pictures, we extract larger connected-components by assigned thresholds. And we perform equal formulations on the height and width of these larger connected-components. If the heights of these larger connected-components are similar and the widths of these larger connected-components are similar, we can get the matching between overlapped pictures and merge them as a bigger picture.
Our matching and merging system has a good performance to our experimental results.

TABLE OF CONTENTS
ABSTRACT IN CHINESE ………………………………. i
ABSTRACT IN ENGLISH ………………………………… iii
ACKNOWLEDGEMENTS …………………………….... v
TABLE OF CONTENTS ………………………………….. vi
LIST OF FIGURES ………………………………………...viii
LIST OF TABLES …………………………………………....x
Chapter 1. Introduction ………………………………………………. 1
1.1 Motivation ……………………………………………………… 1
1.2 Problem Definitions ………………………………………….....4
1.2.1 Overlapping of texts ……………………………………….4
1.2.2 Overlapping of titles/subtitles ……………………………. 5
1.2.3 Overlapping of tables ……………………………….......... 6
1.2.4 Overlapping of pictures …………………………………... 7
1.2.5 Overlapping of mixed type …………………………….......8
1.3 Survey of Related Research …………………………………… 9
1.4 System Description and Assumptions ………………………12
1.4.1 System Description ……………………………………. 12
1.4.2 Assumptions ……………………………………………... 12
1.5 Thesis Organization …………………………………………...14
Chapter 2. Matching and Merging of Plain Text ……………………15
2.1 Overview of Document Layout Analysis …………………… 15
2.1.1 Skew Correction ………………………………………… 16
2.1.2 Connected-Component Extraction ……………………. .19
2.1.3 Connected-Component Classification …………………. 21
2.1.4 Block Segmentation and Classification ………………... 21
2.1.5 Character Segmentation ………………………………... 21
2.2 Connected-Component Based Merging ………………..……..…..…. …. 24
2.2.1 Matching based on connected-components counts …… 28
2.3 Character Based Merging …………………………………………….. 35
2.3.1 Character recognition ……………………………………35
2.3.2 Character based matching ……………………………… 35
2.4 Matching and Merging of Overlapped Titles/Subtitles ………….….…. 40
Chapter 3. Matching and Merging of Non-Text …………………….43
3.1 Matching and Merging of Overlapped Tables ……………….……..……43
3.2 Matching and Merging of Overlapped Pictures ………………..…….….48
3.3 Matching and Merging of Overlapped Mixed Type images ………….....53
Chapter 4. Experimental Results and Analysis ……………………...55
Chapter 5. Conclusions and Future Works ………………………….69
Reference ………………………………………………………………71
LIST OF FIGURES
Fig. 1.1.1 The original image of an input Chinese newspaper document …………3
Fig. 1.2.1 A common overlapping in the right-top of (A) and the down of (B). (A) and (B) are smaller parts of the text images …………………………….4
Fig. 1.2.2 A common overlapping in the right of (A) and the left of (B). (A) and (B) are text images include plain text, title, and subtitle …………………….5
Fig. 1.2.3 A common overlapping in the down of (A) and the top of (B). (A) and (B) are parts of the tables …………………………………………………….6
Fig. 1.2.4 A common overlapping in the down of (A) and the top of (B). (A) and (B) are pictures …………………………………………………..………..…7
Fig. 1.2.5 A common overlapping in the right-down of (A) and the top of (B). (A) and (B) are mixed images including texts, titles, tables, and pictures ….8
Fig. 1.4.1.1 System diagram of matching and merging system of images ………….13
Fig. 2.1.1 (a) The sampling lines in the document image. (b) The pixels in two sampling lines. (c) The skew correction of (a) ……..………………….18
Fig. 2.1.2 (a) Image components are surrounded by text lines. (b) The result of segmentation using connected-component extraction ……..…………..20
Fig. 2.2.1 The conventional method to find overlapping of two images …………24
Fig. 2.2.2 System diagram of matching and merging system of plain text ……….27
Fig. 2.2.1.1 The number of connected-components in a character ………………….28
Fig. 2.2.1.2 Translating a text image into a two-dimensional array that stores the number of connected-components of the text image …………………..29
Fig. 2.2.1.3 (a) Text image segmentation and translation from a text image into an array. (b) The possible matching blocks we choose for matching method.
(c) The detail process of matching method …………………………….33
Fig. 2.3.2 (a) The process of OCR from a text image into an Chinese characters array. (b) The possible matching blocks we choose for matching method. (c) The detail process of matching method …………..……………….39
Fig. 2.4.1 An image with a title and a subtitle …………………………………….41
Fig. 2.4.2 System diagram of matching and merging system of title/subtitle …….42
Fig. 3.1 (a) System diagram of matching and merging system of table ……..…..44
Fig. 3.1.1 (a) A incomplete table (b) A complete table ……………………………45
Fig. 3.1.2 (a) Table that has horizontal and vertical lines. (b) Plain text that has no horizontal and vertical lines ……………………………………………46
Fig. 3.1.3 (a) A table with horizontal and vertical lines (b) A plain text image …………………………………….……………………………..47
Fig. 3.2.1 A mixed type image including pictures ……….………………………..48
Fig. 3.2 (a) System diagram of matching and merging system of picture …...…..49
Fig. 3.2.2 (a) The height and the width of chosen larger connected-component are larger than t1 and are smaller than t2. (b) The matching connected-component with the similar height and similar width. (c) The similarity formulations in detail ……………………………………....52
Fig. 3.3.1 The mixed type image includes text, title, table, and picture ………….54
Fig. 4.1 (a) The smaller size of overlapping and the larger size of the matching block. (b) The numbers of connected-components in characters: (6,1,1,1,1,1,2) …………………………………………………………57
Fig. 4.2 Merging of overlapped text images …………………………………...58
Fig. 4.3 Merging of overlapped texts …………………………………………..61
Fig. 4.4 Merging of overlapped titles/subtitles ……………………………...…62
Fig. 4.5 Merging of overlapped tables …………………………………………63
Fig. 4.6 The extracted picture blocks in image2 are parts of the extracted picture blocks in image1 ………………………………………………….….66
Fig. 4.7 (a) Merging of overlapped pictures. (b) Merging of overlapped pictures ……………………………………..…………………….…..67
Fig. 4.8 Merging of overlapped mixed type images including text, table and picture ………………………………………………………………….68
LIST OF TABLES
Table 4.1 Matching and merging rate of overlapped text images …………………56
Table 4.2 Matching and merging rate of overlapped recognized texts ……………59
Table 4.3 Matching and merging rate of overlapped titles/subtitles ………………59
Table 4.4 Matching and merging rate of overlapped tables ……………………….60
Table 4.5 Matching and merging rate of overlapped pictures …………...………..64
Table 4.6 Matching and merging rate of overlapped mixed type images …………64

REFERENCE
[1] Yi.Lu, “Machine Printed Character Segmentation - an Overview,” Pattern Recognition, Vol. 28, No. 1, pp. 67-80, 1995.
[2] Casey, R.G.; Lecolinet, E, “A survey of methods and strategies in character segmentation” Pattern Analysis and Machine Intelligence, IEEE Transactions on, Vol. 18, No. 7, pp. 690-706, 1996.
[3] R.C. Gonzalez and R.E. Woods, Digital Image Processing, Addision-Wesley publishing company, 1993.
[4] J.R. Parker, Algorithms for Image Processing and Computer Vision, Wiley Computer Publishing.
[5] Jiang, Huei-Fen; Han, Chin-Chuan; Fan, Kuo-Chin, “A fast approach to the detection and correction of skew documents”, Pattern Recognition Letters, Vol. 18, No. 7, pp. 675-686, 1997.
[6] L.G. Brown, “A Survey of Image Registration Techniques”, ACM Computing Surveys, Vol. 24, No. 4, pp. 325-376, 1992.
[7] Dani, Pankaj; Chaudhuri, Subhasis, “Automated assembling of images: image montage preparation”, Pattern Recognition, Vol. 28, No. 3, pp. 431-445, 1995.
[8] Avaninda and Subhasis Chaudhuri, “Robust detection of skew in document images”, IEEE Transaction on Image Processing, Vol. 6, No. 2, 1997.
[9] Inampudi, R.B, “Image mosaicking”, Geoscience and Remote Sensing Symposium Proceedings, IGARSS '98, 1998 IEEE International, Vol. 5, pp. 2363 -2365, 1998.
[10] Ventura, A.D.; Rampini, A.; Schettini, R, “Image registration by recognition of corresponding structures”, Geoscience and Remote Sensing, IEEE Transactions on, Vol. 28, No. 3, pp. 305-314, 1990.
[11] Sawhney, H.S.; Kumar, R, “True multi-image alignment and its application to mosaicing and lens distortion correction”, Pattern Analysis and Machine Intelligence, IEEE Transactions on, Vol. 21, No. 3, pp. 235-243, 1999.
[12] Nakao, T.; Kashitani, A.; Kaneyoshi, A, “Scanning a document with a small camera attached to a mouse”, Applications of Computer Vision, 1998. WACV '98. Proceedings, Fourth IEEE Workshop on, pp. 63-68, 1998.
[13] Nakajima, N.; Tanaka, N.; Yamada, K, “Document reconstruction and recognition from an image sequence”, Pattern Recognition, 1998. Proceedings. Fourteenth International Conference on, Vol. 1, pp. 922-925, 1998.
[14] Fukunaga, K.; Murata, H.; Asano, T.; Izumi, M., “Image registration using an image graph and its application to map matching”, Computers and Digital Techniques, IEE Proceedings E [see also Computers and Digital Techniques, IEE Proceedings-], Vol. 138, No. 2, pp. 79-84, 1991.
[15] Kourogi, M.; Kurata, T.; Hoshino, J.; Muraoka, Y., “Real-time image mosaicing from a video sequence”, Image Processing, 1999, ICIP 99. Proceedings. 1999 International Conference on, Vol. 4, pp. 133-137, 1999.
[16] Guillen Gonzalez, M.; Holifield, P.; Varley, M.R., “Spherical mosaic construction using physical analogy for consistent image alignment”, Image Processing and Its Applications, 1999. Seventh International Conference on (Conf. Publ. No. 465), Vol. 1, pp. 311-315, 1999.
[17] Jun-Wei Hsieh; Hong-Yuan Mark Liao; Kuo-Chin Fan; Ming-Tak Ko, “A fast algorithm for image registration without predetermining correspondences”, Pattern Recognition, 1996., Proceedings of the 13th International Conference on, Vol. 1, pp. 765-769, 1996.
[18] Yihong Gong; Proietti, G.; LaRose, D., “A robust image mosaicing technique capable of creating integrated panoramas”, Information Visualization, 1999. Proceedings. 1999 IEEE International Conference on, pp. 24-29, 1999.
[19] Badra, F.; Qumsieh, A.; Dudek, G., “Rotation and zooming in image mosaicing”, Applications of Computer Vision, 1998. WACV '98. Proceedings., Fourth IEEE Workshop on, pp. 50-55, 1998.
[20] Heung-Yeung Shum and Szeliski, R, “Construction and refinement of panoramic mosaics with global and local alignment”, Computer Vision, 1998. Sixth International Conference on, pp. 953-956, 1998.
[21] Zheng, Q. and Chellappa, R., “A computational vision approach to image registration”, Image Processing, IEEE Transactions on, Vol. 2, No. 3, pp. 311-326, 1993.
[22] T. H. Hildebrandt and W. Liu, “Optical recognition of hand written Chinese characters advances since 1980”, Pattern Recognition, Vol. 26, No. 2, pp. 205-225, 1993.
[23] S.L Yin, “Article Analysis and Title Character Segmentation in Chinese Newspaper”, Department of Computer Science and Information Engineering, College of Electrical Engineering and Computer Science, National Chiao Tung University, 2000.
[24] J.L Chen, “A Study On Structure Extraction, Identification and Data Extraction In Tabular-Form Document Images Processing”, Department of Computer Science and Information Engineering, College of Electrical Engineering and Computer Science, National Chiao Tung University, 2001.

QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top