臺灣博碩士論文加值系統

本篇論文提出了使用資料隱藏的影像竄改回復技術。一靜態影像經方塊式的
小波轉換將它從空間域轉到頻率域後，取出其低頻的係數作為特徵值。再將它嵌
入原圖頻率域的中頻帶，嵌入的資料是用來偵測竄改的區域以及根據偵測出的區
域來回復被竄改的位置。原始影像在經過嵌入資料後仍然具有相當高的影像品
質，人眼幾乎無法分辨嵌入前與嵌入後的圖形有何不同。
當影像散佈在一般公開的環境如網際網路時，難免會遭受到惡意的破壞。因
此當影像遭受到惡意的竄改破壞時，就可以取出之前嵌入的資料以及被竄改破壞
影像的低頻係數來作比對，兩份資料不相同的地方就是被竄改的地方。經由上述
步驟就可以找出被竄改的區域，並且進一步的將被竄改的區域回復。以上偵測和
回復的演算法完全不需要用到其它額外的資訊。
為了加強影像的強韌性，欲嵌入的資料會被備份成三份而且在嵌入之前都會
作隨機排列。隨機排列的作用除了打散外還有避免兩個相同的係數嵌入同一個位
置。實驗結果証明本篇所提出的方法可以抵抗如：切割、消除、替換、塗改和溢
出式填滿的破壞。

This thesis presents an image tamper recovery using data hiding technique. A
source image will be transformed to frequency domain based on blocked discrete
wavelet transform (Blocked-DWT) and the coefficients of the low frequency band are
extracted to be the features. The extracted coefficients are self-embedded into
frequency domain of the source image for tamper detection and recovery. The source
image after embedding still maintains high quality of image. It is very hard to tell the
difference betweens the image before and after embedding.
Images could be maliciously tampered when they spread over Internet. Once the
embedded image is tampered by someone, the tampered area can be located by the
extracted feature. Moreover, the extracted feature is used to recover the tampered area
as well. For robustness, the feature to be embedded is made three copies and
pseudo-random permutation technique before embedding. Pseudo-random
permutation is used to prevent from embedding two identical coefficients in the same
block. Experimental results show that the proposed method is resistant to cropping,
removing, replacement, daub and flood fill.

Chapter 1 Introduction…………………………………………………1
1.1 Motivation...…………………..……………………………………………...1
1.2 The Requirement of Image Authentication…………………………………2
1.3 Thesis Organization…...…………..………………………………….………3
Chapter 2 Background...………………………………………………..4
2.1 What is Steganography?………………………..……………………...……4
2.1.1 Steganography – Introduction……………………………………..… .4
2.1.2 Steganography – Classification...……………………………………..6
2.1.3 Steganography – Application…………………………………………6
2.2 Discrete Wavelet Transform…………………………………………………8
2.2.1 Introduction to Wavelet…………………………………………….… 9
2.2.2 Haar Wavelet……………………………………………………….… 9
Chapter 3 The Proposed Technique of Image Tamper Recovery..… 14
3.1 Data Embedding Algorithm.…………..……………………………………15
3.2 Tamper Detection and Recovery Algorithms……………………………...22
3.2.1 Tamper Detection Algorithm………………………………………..22
3.2.2 Tamper Recovery Algorithm………………………………………..24
Chapter 4 Simulation Results…………………………………………25
4.1 Test Images…………………………………………………………………26
4.2 Test Images with Hiding Information…………………..…………………..27
4.3 Tamper Detection and Recovery…………………………...……………….29
4.4 Comparison with related research results………………………..………….36
Chapter 5 Conclusion………………………………………………….40
References………………………………………………………………41

[1] “Digital watermarking for telltale tamper proofing and authentication” Kundur,
D.; Hatzinakos, D. Proceedings of the IEEE , Volume: 87 Issue: 7 , July 1999
Page(s): 1167 -1180
[2] “An Improved Tamper- Detection Method For Digital Images” Intaek
Kim,Seung-Soo Han and Jung Hwa Shin, Industrial Electronics, 2001. Proceedings.
ISIE 2001. IEEE International Symposium on , Volume: 1 , 2001
[3] “A hybrid watermark for tamper detection in digital images” Fridrich, J. Signal
Processing and Its Applications, 1999. ISSPA '99. Proceedings of the Fifth
International Symposium on , Volume: 1 , 1999 Page(s): 301 -304 vol.1
[4] “Image watermarking for tamper detection” Fridrich , J. Image Processing,
1998. ICIP 98. Proceedings. 1998 International Conference on , Volume: 2 , 1998
Page(s): 404 -408 vol.2
[5] “Digital steganography: hiding data within data” Artz, D. IEEE Internet
Computing, Volume: 5 Issue: 3 , May-June 2001 Page(s): 75 -80
[6] “Information Authentication for a Slippery Hew Age” S. Walton Dr. Dobbs
Journal, vol. 20, no. 4 , pp. 18-26, Apr 1995.
[7] “Fragile Watermarking Using the VW2d Watermark”, R.B. Wolfgang and E.J.
Delp Proceeding of the SPIE/IS&T International Conference on Securith and
Watermarking of Multimedia Contents, January 199, vol. 3657, pp. 204-213.
[8] “Image tamper detection and recovery system based on discrete wavelet
transformation” Kwang-Fu Li; Tung-Shou Chen; Seng-Cheng Wu Communications,
Computers and signal Processing, 2001. PACRIM. 2001 IEEE Pacific Rim Conference
on , Volume: 1 , 2001 Page(s): 164 -167 vol.1
[9] “Information Hiding – A survey” Fabien A. P.Petitcolas, Ross J. Anderson and
Markus G. Kuhn Proceedings of the IEEE, special issue on protection of multimedia
content, 87(7):10621078, July 1999
[10] “Protecting publicly-available images with a visible image watermark” G. W.
Braudaway, K. A. Magerlein, and F. Mintzer, In van Renesse [87], pp. 126-133,
ISBN 0-8194-2033-6
[11] “Making paper - A look into the history of an ancient craft” B. Rudin,.
V¨allingby, Sweden: Rudins, 1990, ISBN 91-970-8882-X,
[12] http://www.research.ibm.com/journal/sj/mit/sectiona/bender.html