臺灣博碩士論文加值系統

本研究提出的資訊隱藏方法是基於像素差值(pixel differencing)的可調式資訊隱藏法。本文提出兩個可調式資訊隱藏法，以下分別稱為Method I與Method II。每個方法皆可分成兩個階段進行機密訊息的藏入。在Method I與Method II兩個方法中的第一階段的步驟是相同的，皆是先將原始影像分組，每一組包含三個像素值。每一像素組的中間像素皆運用LSB 藏入法進行機密訊息的藏入。在第二階段的藏入程序中，Method I以中間像素為基準，分別與其兩側的像素進行比對以創立出兩個虛擬像素，並將虛擬像素值與兩側像素值進行分組。藉由改變虛擬像素值來放大或縮小虛擬像素與左右兩側像素的差距，讓差距落在較大或較小的區間，藉此提高藏入量或提高影像品質，再進行像素值差異擴張法(pixel-value differencing; 以下簡稱PVD)的運算將機密訊息藏入於兩側像素中。此方法的目的在於可以依照使用者需求來改變虛擬像素值以提高藏入量亦或是提高影像品值。另外，Method II的第二階段首先創建2個參考矩陣，接著將每一像素組的兩側像素與中間像素進行差距計算，並以此差距是否在門檻值以內或外來選定參考矩陣。以兩側像素為座標在256×256平面上找出參考矩陣中符合機密訊息的座標作為偽裝影像的像素對。此方法亦可依需求調整參考矩陣和門檻值得到所需的藏入量和影像品質。
Method I可擴增(extension)與縮減(shrinking)資訊藏量，以維護影像品質。其效能與 Khodaei et al.學者所提出的方法相比藏量的增加約有0.11bpp(約為28836bits)與影像品質PSNR值提昇1.3dB，但是Method I概念相當的簡單，可應用的範圍很廣，可針對各種PVD方法進行改良。Method II在藏入量在3bpp時影像品質依然可維持在40dB以上，如果再進行調整，有張影像可以達到4.17bpp，影像品質為30.25dB，影像品質在30dB以上人的視覺是看不出影像的變化。

The steganographic method for data hiding proposed in this thesis is related to adjustable information hiding methods based on pixel differencing. We propose two adjustable information hiding methods, called Method I and Method II. Each method can be divided into two stages to hide confidential information. The steps in the first stage of Method I and Method II are same; the original images are grouped, and each group contains three pixels. The intermediate pixels of each group all use a LSB hiding method to hide confidential information. In the second stage of information hiding, Method I is based on intermediate pixels compared with the pixels on both sides respectively to create two virtual pixels. Each virtual pixel is grouped together with its corresponding side pixel. The gap between the virtual pixel and the side pixel is enlarged or reduced by changing the values of virtual pixels. The gap can be fine-tuning adjusted by the virtual pixels to a larger interval to increase the hiding amount or to improve the image quality at the area with the smaller interval difference. And the pixel-value differencing (PVD) is then used to store the confidential information in the pixels of both sides. The purpose of this method is to change the values of virtual pixels according to the user's requirement to increase the hiding amount or increase the image quality. In addition, the second stage of Method II first creates two reference matrices, and then calculates the gap between the pixels on both sides of each pixel group and the intermediate pixels; and the reference matrix is selected based on whether the gap is within or outside the threshold value. Based on both sides of pixels as coordinates in the 256×256 plane, the value at the coordinates of reference matrix that matches the confidential digit is treated as the coordinates of a pair of pixels for camouflage images. This method can also adjust the reference matrix and threshold to obtain the required hiding amount and image quality.
Method I also can have two possibilities which extend or shrink the embedding capacity to maintain image quality. Comparing with the method proposed by Khodaei and Faez, its performance increases in the amount of hiding about 0.11bpp (about 28836bits) and PSNR of image quality by 1.3dB. However, the concept of Method I is fairly simple and can be applied in a wide range of ways, and it can improve for various PVD methods. In the amount of hiding about 3 bpp, the image quality of Method II can still be maintained at 40 dB or more. If you make adjustments, there are images which can reach the amount of hiding about 4.17 bpp, and the image quality of 30.25 dB, where for image quality in more than 30 dB, human vision cannot see the image changes.

目錄
中文摘要........................................................................................................................ I
Abstract ....................................................................................................................... III
誌謝............................................................................................................................... V
目錄............................................................................................................................ VII
表目錄.......................................................................................................................... IX
圖目錄........................................................................................................................... X
第 1 章 緒論 ....................................................................................................... 1
1.1 研究背景與動機.................................................................................... 1
1.2 研究動機及目的.................................................................................... 2
1.3 論文架構................................................................................................ 6
第 2 章 文獻探討 ............................................................................................... 7
2.1 符號定義................................................................................................ 7
2.2 像素值差異擴張技術(PVD)方法回顧 ................................................. 9
2.2.1 藏入步驟........................................................................................ 9
2.2.2 取出機密訊息.............................................................................. 11
2.2.3 PVD 的藏入與取出範例說明..................................................... 12
2.3 Wu et al.學者的 PVD 與 LSB 的藏入法 ............................................ 14
2.3.1 藏入步驟...................................................................................... 14
2.3.2 機密訊息取出.............................................................................. 16
2.3.3 Wu et al.學者的 PVD 與 LSB 的藏入法的範例說明 ................ 18
2.4 可調式的 PVD 與 LSB 藏入法 .......................................................... 20
2.4.1 藏入步驟...................................................................................... 20
2.4.2 機密訊息取出.............................................................................. 23
2.4.3 可調式的 PVD 與 LSB 藏入法藏入與取出範例說明 .............. 25
2.5 基於 PVD 與 EMD 的藏入法 ............................................................. 27
2.5.1 藏入步驟...................................................................................... 27
2.5.2 機密訊息取出.............................................................................. 30
2.5.3 基於 PVD 與 EMD 的藏入法藏入與取出範例說明 ................. 31
2.6 Jung 學者的基於 PVD 與 LSB 增大容量的方法 .............................. 33
2.6.1 藏入步驟...................................................................................... 33
2.6.2 機密訊息取出.............................................................................. 36
2.6.3 Jung 學者的基於 PVD 與 LSB 增大容量的方法藏入與取出範例說明 ............................................................................................. 38
2.7 運用差值對吻合之 PVD 藏入法........................................................ 41
2.7.1 藏入步驟...................................................................................... 41
2.7.2 機密訊息取出.............................................................................. 42
2.7.3 運用差值對吻合之 PVD 藏入法藏入與取出範例說明............ 42
第 3 章 植基於像素差值之可調式資訊隱藏法 ............................................. 46
3.1 植基於最不重要位元取代法及像素值差異擴張法之影像隱寫術.. 46
3.1.1 方法介紹與藏入步驟.................................................................. 47
3.1.2 藏入機密訊息程序...................................................................... 48
3.1.3 取出機密訊息程序...................................................................... 51
3.1.4 放大或縮小差距的使用時機及差異.......................................... 53
3.1.5 範例說明...................................................................................... 54
3.2 運用像素對吻合及像素值差異技術之可調式資訊隱藏法.............. 58
3.2.1 方法介紹與藏入步驟.................................................................. 58
3.2.2 藏入機密訊息程序...................................................................... 60
3.2.3 取出機密訊息程序...................................................................... 62
3.2.4 範例說明...................................................................................... 63
第 4 章 實驗結果分析與討論 ......................................................................... 67
4.1 實驗環境與評定準則.......................................................................... 67
4.2 運用最不重要位元取代及像素值差異擴張技術之可調式資訊隱藏法 ......................................................................................................... 70
4.3 運用像素對吻合及像素值差異技術之可調式資訊隱藏法.............. 75
4.4 方法比較.............................................................................................. 97
第 5 章 結論與未來展望 ............................................................................... 102
5.1 結論.................................................................................................... 102
5.2 未來展望............................................................................................ 104
參考文獻.................................................................................................................... 105

表目錄
表 2-1 區間量表 ........................................................................................ 11
表 2-2 區間量表(門檻值T0=15) ............................................................ 16
表 2-3 區間量表(門檻值T0=31) ............................................................ 16
表 2-4 區間量表 I ....................................................................................... 23
表 2-5 區間量表 II...................................................................................... 23
表 2-6 區間量表 ......................................................................................... 30
表 2-7 區間量表(門檻值T0=31) ............................................................. 36
表 3-1 區間量表 ......................................................................................... 51
表 3-2 參考矩陣大小與藏量 .................................................................... 62
表 4-1 Method I 的區間量表 ..................................................................... 70
表 4-2 Khodaei et al.學者的方法[2]與 Method I 的比較表 ..................... 72
表 4-3 Khodaei et al.學者的方法[2]與 Method I 的差值表 ..................... 72
表 4-4 Method I與 Khodaei et al.學者的方法在相同藏量的 PSNR 比較表
.............................................................................................................. 73
表 4-5 參考矩陣大小與藏量 ..................................................................... 75
表 4-6 Method II 修改門檻值的 bpp 與 PSNR 比較表(l=2,u=3) .... 76
表 4-7 Method II 修改門檻值參考矩陣使用率表 .................................... 77
表 4-8 Method II 修改參考矩陣的 bpp 與 PSNR 比較表(T0=8) ......... 82
表 4-9 Method II 修改RTu參考矩陣的 bpp 與 PSNR 比較表(T0=128、u=10)
.................................................................................... 84
表 4-10 Method II 修改參考矩陣的 bpp 與 PSNR 比較表 ...................... 88
表 4-11 Method II 各門檻值與參考矩陣的 bpp 與 PSNR 比較表 .......... 93
表 4-12 Method II 與 Jeanne 學者的方法[8]在相同條件下的比較表 ..... 95
表 4-13 Method II 與 Jeanne et al.學者的方法[8]在相同條件下的差值表
.............................................................................................................. 96
表 4-14 Method II 與 Jeanne et al.學者的方法[8]的比較表 ..................... 96
表 4-15 資訊負載量 bpp 比較表 .............................................................. 97
表 4-16 影像信號雜訊比 PSNR(單位 dB)比較表 ................................... 98
表 4-17 相同藏量下 Jung et al.學者的方法[3]與 Method I 與 Method II
的比較表.............................................................................................. 99
表 4-18 相同藏量下 Jeanne et al.學者的方法[8]與 Method I 與 Method II
的比較表............................................................................................ 100
表 4-19 相同藏量下 Shen et al.學者的方法[11]與 Method I 與 Method II
的比較表............................................................................................ 101
X
圖目錄
圖 2-1 PVD 方法範例流程圖 .................................................................... 13
圖 2-2 Wu et al.學者的 PVD 與 LSB 的藏入法範例流程圖.................... 19
圖 2-3 可調式的 PVD 與 LSB 藏入法範例流程圖 ................................. 26
圖 2-4 基於 PVD 與 EMD 的藏入法範例流程圖 ................................... 32
圖 2-5 Jung 學者的基於 PVD 與 LSB 增大容量的方法範例一流程圖 . 40
圖 2-6 Jung 學者的基於 PVD 與 LSB 增大容量的方法範例二流程圖 . 40
圖 2-7 RTl、RTu參考矩陣 ........................................................................ 44
圖 2-8 2×2 像素區塊 .................................................................................. 44
圖 2-9 RTu參考矩陣藏入範例 .................................................................. 45
圖 2-10 2×2 像素區塊 ................................................................................ 45
圖 2-11 RTu參考矩陣取出範例 ................................................................ 45
圖 3-1 基於最不重要位元取代法及像素值差異擴張法之影像隱寫術之流程圖 ............................................................................................. 47
圖 3-2 一個群組像素 ................................................................................ 56
圖 3-3 第一階段將 3 個 bit 藏到中間像素值 .......................................... 57
圖 3-4 第二階段藏入時所創建虛擬像素分別為 54 及 62 ..................... 57
圖 3-5 偽裝像素 ........................................................................................ 57
圖 3-6 第二階段取出訊息時創建虛擬像素分別為 54 及 62 ................. 57
圖 3-7 基於最不重要位元取代法及像素值差異擴張法之影像隱寫術之
範例流程圖 ..................................................................................... 57
圖 3-8 運用像素對吻合及像素值差異技術之可調式資訊隱藏法之流程
圖 ..................................................................................................... 58
圖 3-9 RTl與RTu參考矩陣範例 ................................................................ 62
圖 3-10 RTl與RTu參考矩陣例子 .............................................................. 65
圖 3-11 RTu參考矩陣 ................................................................................ 65
圖 3-12 運用像素對吻合及像素值差異技術之可調式資訊隱藏法之範例流程圖 ......................................................................................... 66
圖 4-1 實驗測試影像 ................................................................................. 69
圖 4-2 Method I與 Khodaei et al.學者的方法在相同藏量的 PSNR 折線圖
(Baboon 影像) ................................................................................. 74
圖 4-3 (a)Lena, (b)Baboon, (c)Peppers 修改門檻值的使用率折線圖...... 78
圖 4-4 (d)Boat, (e)Peppers, (f)Boat 修改門檻值的使用率折線圖 ........... 79
圖 4-5 (g)Livingroom, (h)Stream_and_Bridge, (i)Pirate修改門檻值的使用
率折線圖 ......................................................................................... 80
XI
圖 4-6 (j)Fishing_Boat, (k)Elaine, (l)House 修改門檻值的使用率折線圖
......................................................................................................... 81
圖 4-7 (a)Lena, (b)Baboon, (c)Peppers, (d)Boat, (e)Airplane, (f)Tiffany 修
改 RTl 參考矩陣的 PSNR 折線圖 .................................................. 85
圖 4-8 (g)Livingroom, (g)Stream_and_Bridge, (i)Pirate, (j)Fishing_Boat,
(k)Elaine, (l)House 修改 RTl 參考矩陣的 PSNR 折線圖 .............. 86
圖 4-9 (a)Lena, (b)Baboon, (c)Peppers 修改 RTu參考矩陣的 PSNR 與 bpp
折線圖 ............................................................................................. 89
圖 4-10 (d)Boat, (e)Airplane, (f)Tiffany修改RTu參考矩陣的PSNR與bpp
折線圖 ............................................................................................. 90
圖 4-11 (g)Livingroom, (h)Stream_and_Bridge, (i)Pirate 修改 RTu 參考矩
陣的 PSNR 與 bpp 折線圖 ............................................................. 91
圖 4-12 (j)Fishing_Boat, (k)Elaine, (l)House 修改 RTu 參考矩陣的 PSNR
與 bpp 折線圖 ................................................................................. 92
圖 4-13 Method II 各影像門檻值與參考矩陣的 PSNR 折線圖 .............. 94
圖 4-14 Method II 各影像各門檻值與參考矩陣的 bpp 折線圖 .............. 94

參考文獻
[1]C.C. Chang, J.Y. Hsiao, and C.S. Chan, “Finding optimal least-significant-bit substitution in image hiding by dynamic programming strategy,” Pattern Recognit., vol. 36, no. 7, pp. 1583–1595, 2003.
[2]M. Khodaei and K. Faez, “New adaptive steganographic method using least-significant-bit substitution and pixel-value differencing,” IET Image Process., vol. 6, no. September, pp. 677, 2012.
[3]K.-H.Jung, “High-capacity steganographic method based on division arithmetic and LSB replacement,” Imaging Sci. J., vol. 180, no. August, pp. 213–221, 2012.
[4]S.Y. Shen and L.H. Huang, “A data hiding scheme using pixel value differencing and improving exploiting modification directions,” Comput. Secur., vol. 48, pp. 131–141, 2015.
[5]D.C. Wu and W.H. Tsai, “A steganographic method for images by pixel-value differencing,” Pattern Recognit. Lett., vol. 24, no. 9–10, pp. 1613–1626, 2003.
[6]H.C. Wu, N.I. Wu, C.S. Tsai, and M.S. Hwang, “Image steganographic scheme based on pixel-value differencing and LSB replacement methods,” Chinese J. Electron., vol. 14, no. 1, pp. 115–118, 2005.
[7]X. Zhang and S. Wang, “Efficient steganographic embedding by exploiting modification direction,” IEEE Commun. Lett., vol. 10, no. 11, pp. 781–783, 2006.
[8]J. Chen, “A PVD-based data hiding method with histogram preserving using pixel pair matching,” Signal Process. Image Commun., vol. 29, no. 3, pp. 375–384, 2014.
[9]C.K. Chan and L.M. Cheng, “Hiding data in images by simple LSB substitution,” Pattern Recognit., vol. 37, no. 3, pp. 469–474, 2004.
[10]W. Hong and T.S. Chen, “A novel data embedding method using adaptive pixel pair matching,” IEEE Trans. Inf. Forensics Secur., vol. 7, no. 1, pp. 176–184, 2012.
[11]T.D. Kieu and C.C. Chang, “A steganographic scheme by fully exploiting modification directions,” Expert Syst. Appl., vol. 38, no. 8, pp. 10648–10657, 2011.
[12]C.H. Yang, C.Y. Weng, H.K. Tso, and S.J. Wang, “A data hiding scheme using the varieties of pixel-value differencing in multimedia images,” J. Syst. Softw., vol. 84, no. 4, pp. 669–678, 2011.
[13]V. Sabeti, S. Samavi, M. Mahdavi, and S. Shirani, “Steganalysis of pixel-value differencing steganographic method,” IEEE Pacific RIM Conf. Commun. Comput. Signal Process. - Proc., no. September, pp. 292–295, 2007.
[14]C.C. Chang and H. Tseng, “A steganographic method for digital images using side match,” Pattern Recognit. Lett., vol. 25, no. 12, pp. 1431–1437, 2004.
[15]G. Swain, “Digital image steganography using nine-pixel differencing and modified LSB substitution,” Indian J. Sci. Technol., vol. 7, no. 9, pp. 1444–1450, 2014.
[16]K.C. Chang, C.P. Chang, P.S. Huang, and T.M. Tua, “A novel Image Steganographic Method Using Tri-way Pixel-Value Differencing,” J. Multimed., vol. 3, no. 2, pp. 37–44, 2008.
[17]X. Li, J. Li, B. Li, and B. Yang, “High-fidelity reversible data hiding scheme based on pixel-value-ordering and prediction-error expansion,” Signal Processing, vol. 93, no. 1, pp. 198–205, 2013.
[18]Y.Q. Shi, “Reversible data hiding,” vol. 16, no. 3, pp. 1–12, 2005.
[19]J. Tian, “Reversible data embedding using a difference expansion,” IEEE Trans. Circuits Syst. Video Technol., vol. 13, no. 8, pp. 890–896, 2003.
[20] C.H. Yang “Inverted pattern approach to improve image quality of information hiding by LSB substitution,” Pattern Recognit., vol. 41, no. 8, pp. 2674–2683, 2008.
[21] SIPI Image Database, (http://sipi.usc.edu/database/database.php)
[22] ImageProcessingPlace, (http://www.imageprocessingplace.com/)