臺灣博碩士論文加值系統

中 文 摘 要
隨著多媒體廣泛的成長與普及，各式新聞、雜誌與網頁都常出現於我們的日常生活中，然而，在各種網路上所下載的資料中，文字顯然在這些資料的瞭解性中伴演著重要角色。除此之外，為了使各個不同母語的人可以與這些資料溝通無礙，文件中的文字偵測與語言翻譯的重要的性於是相形增加，當然，一個完美的語言翻譯即是基於一個好的文字偵測系統。
因此，本論文為了達到完美的語言翻譯，我們提出了一個新的文字偵測技術，可達到一個極低的假警報率(false alarm rate)。首先，以類神經網路的彩色量化法使得顏色類似的文字可被量化成相同的顏色，接著我們使用三維的統計長條圖分析法(3D histogram analysis)來選擇幾個可能的文字候選色，如此，對於每個可能的文字候選色我們可以分別萃取出它們的相對雙色調圖(binary images)，之後我們使用相連物件分析法(connected component analysis)與兩個型態學的運算子(morphological operator)於每一張雙色調圖來找出可能的文字區域(text region)，最後，我們使用高斯的拉普拉斯邊緣檢測器(L.O.G edge detector)來對可能的文字區域做更進一步的確認，同時，多層量化(multi-layer color quantization)的技術可以讓我們大大的降低假警報率。
除了彩色文件中的文字偵測，我們也將我們提出的文字偵測法應用於彩色文件 (同時擁有圖片及文字在其中的圖)的壓縮，例如報紙、雜誌等。此壓縮法可以使得壓縮率提高，再者，當低位元傳送時，文字可以保持它的清析度使人眼看得較為清楚。

Abstract
As the growth of multimedia components, News, magazines, Web pages, etc are everywhere in our life. However, text in these documents plays an important role when people need to realize details of their downloaded data. Besides that, in order to make people who speak different languages know content of these data easily, text localization and translation in color images are becoming more and more important, it is clear that good text translation can be achieved if we can accurately localize text regions.
In order to achieve good translating performance, we propose a novel approach to detect text in color images with very low false alarm rate. First of all, neural network color quantization is used to compact text color. Second, 3D histogram analysis chooses several colors candidates, and then extracted each of these color candidates to obtain several bi-level images. For each extracted bi-level image, connected component analysis and several morphological operators are fed to hold some boxes that are possible text regions. At last, we can use L.O.G edge detector to authenticate accurate text regions from each possible text regions. Meanwhile, in complex color images, multi-quantization layers can be integrated to reject non-text parts and reduce false alarm rate.
In addition to localize text regions in color images, we can also apply the text localization technique in the compression of compound documents such as magazines and newspaper. The application can not only reduce transmitting rate effectively but also hold text part clear when low bit-rate transmitting.

Contents
Chapter 1 Introduction…………………………………………………………………….1
Chapter 2 Review of Some Conventional Text Detection Technique in Color Image…5
2.1 Introduction……………………………………………………………………………...6
2.2 Some Text Detection Algorithms in Related Works…………………………………….7
2.2.1 Multi-Resolution Layer Method……………………………………………….8
2.2.2 Spatial Variance Method……………………………………………………..10
2.2.3 Edge filtering Method………………………………………………………..12
2.2.4 Color Quantization Method………………………………………………….17
2.3 Discussions……………………………………………………………………………..19
Chapter 3 Multi-Layer Color Quantization Text Detection in Complex Color Images……………………………………………………………………………………….25
3.1 Introduction…………………………………………………………………………….25
3.2 The Proposed Multi-Layer Color Quantization Algorithm………………………….…28
3.2.1 SOFM Neural Network Color Quantization…………………………………28
3.2.2 3D Histogram Analysis………………………………………………………35
3.2.2 Morphological Operating……………………………………………………39
3.2.4 Connected Component Analysis……………………………………………...42
3.2.5 Multi-Layer Combination……………………………………………………47
3.3 Experiment Results…………………………………………………………………….55
3.4 Comparisons and discussions………………………………………………………….58
Chapter 4 Compound Document Compression Based on Text Detection Technique...61
4.1 Introduction……………………………………………….……………………………62
4.2 Mixed Raster Content (MRC) Model…………………….……………………………64
4.3 The Proposed Decomposition Method in MRC………………………………………..69
4.3.1 The Proposed Segmentation Procedure……………………………………...70
4.3.2 An Overview of JPEG2000…………………………………………………..77
4.3.3 An Overview of JBIG………………………………………………………...83
4.3.4 Bitstream Organization and Decoding Procedure…………………………...87
4.4 Experiment Results and discussions…………………………………………………...87
Chapter 5 Conclusion and Future Works……………………………………………….93
5.1 Conclusion……………………………………………………………………………..93
5.2 Future Works…………………………………………………………………………...95
Reference……………………………………………………………………………………97

References
[1] R.Lienhart, and A. Wernicle “Localizing and Segmenting text in Images and Videos”, IEEE Trans. Circuits and Systems for Video Technology, pp.256-268, Apr.2002
[2] Robert M. Haralick, and Linda G. Shapiro “Computer and Robot Vision” vol. 1, Addison Wesley, 10877
[3] Victor Wu, Raghavan Manmatha, and Edward M. Riseman “TextFinder: An Automatic System to Detect and Recognize Text In Images”, IEEE Trans. on Pattern Analysis and Machine Intelligence, vol.21, no. 11, pp.1223-1229, Nov.1999
[4] Huiping Li, David Doermann, and Omid Kia “Automatic Text Detection and Tracking in Digital Video”, IEEE Trans. on Image Processing, vol.9, no. 1 pp.156, Jan.2000
[5] Yu Zhong, Kalle Karu, and Anil K. Jain “Locating Text in Complex Color Images”, Pattern Recognition, 28:1523-1535, pp. 146-149, 1995
[6] Anil K. Jain, and Bin Yu “Automatic Text Location in Images and Video Frames”, Pattern Recognition, vol. 31, no. 12, pp. 2055-2076, 1988.
[7] J. Gao, and J. Yang “An Adaptive Algorithm for Text Detection from Natural Secnes”, Proceedings of Computer Vision and Pattern Recognition (CVPR), pp. 84-89, 2001
[8] Jianming Hu, Jie Xi, and Lide Wu “Automatic Detection And Verification of Text Regions in News Video Frames”, International Journal of Pattern Recognition and Artificial Intelligence, vol. 16, no. 2, pp. 257-271, 2002
[9] Jong Ryul Kim and Young Shik Moon “Extraction of Text Regions and Recognition of Characters from Video Inputs”, PCM 2002, LNCS 2532, pp. 767-774, 2002
[10] M. Cai, J. Song, and M. R. Lyu “A New Approach for Video Text Detection”, IEEE Intl. Conf. Image Processing, pp. 117-120, 2002
[11] Y. Yang, X. Chen, J. Zhang, Y. Zhang, and A. Waibel “Automatic Detection and Translation of Text from Natural Scenes”, IEEE, Intl. Conf. Acoustics, Speech, and Signal Processing (ICASSP), pp. 2101-2104, May, 2002
[12] William K. Pratt “Digital Image Processing” third edition, Wiley Interscience.
[13] S.C. Pei, and Y.S. Lo “Color Image Compression and Limited Display Using Self-Organization Kohonen Map”, IEEE, Trans. On Circuits and System for Video Technology, Vol. 8, no. 2, pp. 191-205, Apr 1998.
[14] S. Prabhakar, H. Cheng, John C. Handley, Z. Fan and Y. W. Lin “Picture-Graphics Color Image Classification”, IEEE, ICIP, pp. 785-788.
[15] K. Sobottka, H. Bunke, and H. Kronenberg “Identification of Text on Colored Book and Journal Covers”, Document Analysis and Recognition, 1999. ICDAR '99. Proceedings of the Fifth Intl. Conf., 20-22 Sep. 1999, pp. 57 -62
[16] Standardization of group 3 facsimile apparatus for document transmission, ITU-T Rec. T.4, July 1996.
[17] Facsimile coding schemes and coding control functions for group 4 facsimile apparatus, ITU-T Rec. T.6, Nov 1988
[18] Information Technology — Coded representation of picture and audio information — Progressive bi-level image compression, ITU-T Rec. T.82, Mar. 1995
[19] ISO/IECJTC1/SC29 JBIG Comm., http://www.jpeg.org/public/jbigpt2.htm, Aug. 21, 1988.
[20] W. P. Pennebaker and J. L. Mitchell, JPEG: Still Image Compression Standard: Van Nostrand Reinhold, 1993
[21] JPEG 2000 Committee, Working Draft 2.0, ISO/IEC JTC1/SC29 WG1, June 25, 1999
[22] JPEG 2000 Committee, JPEG 2000 Verification Model (Technical Description), ISO/IEC JTC1/SC29 WG1, Apr. 22, 1999
[23] H. T. Fung and K. J. Parker, “Segmentation of scanned documents for efficient compression”, in Proc. SPIE Vis. Commun. Image Processing, vol. 2727, Orlando, FL, 1996, pp. 701-712]
[24] D. Huttenlocher and W. Rucklidge, “Digipaper: Aversatile color document image representation”, in Proc. IEEE Intl. Conf. Image Proc., Kobe, Japan, Oct. 1999
[25] L. Bottou, P. Haffner, P. Howard, P. Simard, Y. Bengio, and Y. LeCun, “High quality document image compression using DjVu”, J. Electron. Imag., vol. 7, pp. 410-425, July 1998
[26] L. Bottou, P. Haffner, P. Howard, and Y. LeCun, “Color document on the Web with DjVu”, in Proc. IEEE Int. Conf. Image Processing, Kobe, Japan, Oct. 1999
[27] H. Cheng and C. Bouman, “Document compression based on multiscale segmentation”, in Proc. IEEE Int. Conf. Image Processing, Kobe, Japan, Oct. 1999, 25PS1.8
[28] A. Said and A. Drukarev, “Simplified segmentation for compound image compression”, in Proc. IEEE Int. Conf. Image Processing, Kobe, Japan, Oct. 1999, 25PS1.5
[29] File format for internet fax, ftp://ftp.isi.edu/in-notes/rfc2301.txt, L. McIntyre, S. Zilles, R. Buckley, D. Venable, G. Parsons, and J. Rafferty, Eds., Mar. 1998
[30] Mixed Raster Content ITU-T Study Group 8, Question 5, Draft Recommendation T.44, May 1997
[31] J. Huang, Y. Wang, and E. Wong, “Check image compression using a layered coding method”, J. Electron. Image, vol. 7, pp. 426-442, July 1998.
[32] R. de Queiroz, R. Buckley, and M. Xu, “Mixed Raster Content (MRC) model for compound image compression”, in Proc. Int. Conf. Image Processing, vol. 3653, Feb. 1999, pp. 1106-1117
[33] R. L. de Queiroz, Z. Fan, T. D. Tran, “Optimizing block-thresholding segmentation for multilayer compression of compound images”, IEEE Trans. Image Proc., vol. 9, no. 9, Sept 2000
[34] J. M. Shapiro “Embedded image coding using zero trees of wavelet coefficients ”, IEEE Trans. Signal Process. 41 (12), Dec. 1993, pp. 3445-3462
[35] A. Said, W. A. Pearlman “A new fast and efficient image codec based on set partitioning in hierarchical trees”, IEEE Trans. Circuits Systems Video Technol. 6 (3) Jun. 1996, pp. 243-250