數位媒體偽裝術與可逆式資訊隱藏技術之研究

The Study of Steganography and Reversible Data Hiding Techniques for Digital Media

陳星琳(Hsing-Ling Chen)

指導教授 ： 李金鳳

朝陽科技大學/資訊學院/資訊科技研究所博士班/博士(2011年)

摘要

針對使用數位媒體進行祕密傳輸的主題，本論文提出在空間域中兩種不同類型的資訊隱藏方法分別為：不可逆式資訊隱藏與可逆式資訊隱藏。在不可逆式資訊隱藏的方面，為了提升嵌入或取出的執行效能與減少記憶體的需求量，一個植基於模數函數的資訊隱藏方法被提出，由於機密資料是藉由模數函數操作來進行嵌入或取出的處理，因此一個好的執行效能可被實現且所需的記憶體也相當的小，此外利用像素群組與變型的卡笛生乘積的組合來提升所嵌入資料的安全性。
　　在可逆式資訊隱藏的方面，一個植基於差值擴張法之可逆式資訊隱藏的方法首先被提出來，此方法是延伸於Tian學者所提出的差值擴張法，然而本論文方法在嵌入與取出的過程中不需要位置地圖的指引，因此本論文方法的貢獻在於沒有任何高效率的無失真壓縮法的協助下，亦能有效提供高資訊隱藏量(將近1bpp)且還能保有不錯的影像品質。
　　接著本論文提出另一個植基於直方圖法之可逆式資訊隱藏技術，雖然植基於直方圖法之可逆式資訊隱藏技術可獲得很高的影像品質，但其藏量是受限制的。為了能提升資訊隱藏量與降低影像失真，本論文方法採用預測差值來取代原始像素值去攜帶機密資料，在此預測差值是來自於原始影像與預測影像之間的像素差值，由於預測影像與原始影像十分相似，因此所獲得的預測差值將會趨近於0。換言之，大量的峰值點將趨近於0的附近，故本論文方法充分利用此特性來提升資訊隱藏量與降低失真。當達高藏量的需求時，僅有一個門檻值需被記錄而非需記錄大量之峰值點與零值點資訊，此外多層次嵌入的技術也有助於達到提升資訊隱藏量之效。
　　最後本論文提出一個利用變型霍夫曼編碼的資訊隱藏方法，在該方法中機密資料是被嵌入於壓縮碼中，本論文方法主要目的是利用文字檔案來進行機密訊息溝通、提供高資訊隱藏量、提升所嵌入之機密資料的安全性與減少傳輸的成本。在本論文方法中每一個變型霍夫曼樹的葉子至少可以用來攜帶一位元的機密資料。此外利用機密金鑰來產生每一個變型霍夫曼樹的葉子之偽裝壓縮碼，以保護所嵌入機密資料的安全。
　　因為高藏量與高影像品質或高藏量與高壓縮率是無法同時被達成的，因此本論文中所有方法的機密資料嵌入量皆俱備有調性，機密資料嵌入量可視實際應用需求予以彈性的調整。當較低量的機密資料被嵌入時，可以獲得較高的影像品質或壓縮率。反之在一個仍可接受的影像品質或壓縮率的條件下，高資訊藏量可被實現。

關鍵字

文字資訊隱藏 ； 可逆式資訊隱藏 ； 資訊隱藏 ； 隱藏學 ； 無失損壓縮

並列摘要

For a covert transmission subject using digital media, in the spatial domain, this thesis proposes two kinds of data hiding schemes, i.e. irreversible data hiding and reversible data hiding. As for irreversible data hiding, a scheme based on modulus function is developed to enhance the performance of the embedding/extracting process and reduce the memory requirement. The confidential data are embedded/extracted using modulus function operation, ensuring a satisfactory performance and extremely small memory requirement. Moreover, the pixel group and the variant Cartesian product are exploited to enhance the security of embedded data.
　　As for reversible data hiding, a scheme based on prediction-error expansion is first proposed. The proposed scheme is an extension of the difference expansion method developed by Tian. However, the proposed scheme does not require the location map, which markedly influences the amount of confidential data that can be transmitted. Therefore, the proposed scheme has a high embedding capacity that can achieve an approximately 1 bpp with an acceptable visual effect without assistance from an efficient lossless compression algorithm.
　　Next, this thesis shows a histogram-based reversible data hiding scheme. Although common histogram-based reversible data hiding schemes can achieve high image quality, embedding capacity is restricted. To improve embedding capacity and retain low distortion, the proposed scheme uses prediction-error values, which are derived from the difference between an original image and a predictive image, instead of using the original pixels to convey a secret message. Since the obtained predictive image is very similar to the original image, prediction-error values are to be tended to zero. That is, a great quantity of peak points gathers around zero. The proposed scheme takes full advantage of this property to increase embedding capacity and retain slight distortion. Only a threshold is needed to record, not a large amount of information of peak and zero points, when high embedding capacity is required. Additionally, a multilevel mechanism is helpful for further increase embedding capacity.
　　Finally, this thesis presents a data scheme using lossless compression coding called variable Huffman coding. A secret message is embedded into compression codes. The goals of the proposed scheme are to covert communication in text files, provide high embedding capacity, improve the security of embedded secret messages, and reduce transmission cost. Each leaf in the variable Huffman tree in the proposed scheme can be utilized to convey at minimum one secret bit.Additionally, secret keys are utilized to generate the stego-compression code for each leaf on the variable Huffman tree to protect the embedded message.
　　The embedding capacity of all proposed schemes is devised scalable because a high embedding capacity and a high image quality or a high embedding capacity and a high compressed rate cannot meet at the same time. The embedding capacity of proposed schemes is adjusted according to actual applications. A higher image quality or higher compressed rate is produced when a lower amount of confidential data are embedded. On the contrary, a higher capacity is achieved with an acceptable image quality or compressed rate.

並列關鍵字

Reversible data hiding ； Steganography ； Data hiding ； Text data hiding ； Lossless compression

參考文獻

[1]A. M. Alattar (2004), “Reversible watermark using the difference expansion of a generalized integer transform,” IEEE Transactions on Image Processing, Vol. 13, No. 8, pp. 1147-1156.

[5]M. U. Celik, G. Sharma, A. M. Tekalp and E. Saber (2005), “Lossless generalized-LSB data embedding,” IEEE Transactions on Image Processing, Vol. 14, No. 2, pp. 253-266.

[6]C. K. Chan and L. M. Cheng (2004), “Hiding data in images by simple LSB substitution,” Pattern Recognition, Vol. 37, No. 3, pp. 469-474.

[7]C. C. Chang, C. F. Lee and L. Y. Chuang (2009), “Embedding secret binary message using locally adaptive data compression coding,” Journal of Computer Sciences and Engineering Systems, Vol. 3, No. 1, pp. 55-61.

[8]C. C. Chang and P. Y. Lin (2008), “Adaptive watermark mechanism for rightful ownership protection,” Journal of Systems and Software, Vol. 81, No. 7, pp. 1118-1129.