可逆式資訊隱藏方法使用邊緣敏感度分析、不對稱直方圖及雙影像技術

Reversible Data Hiding Schemes Using Edge-Sensing Analysis, Asymmetric-Histogram and Dual Image Technology

曾俊雅(Chun-Ya Tseng)

指導教授 ： 呂慈純

朝陽科技大學/資訊學院/資訊管理系/博士(2015年)

摘要

可逆式資訊隱藏技術最常應使用直方圖位移策略進行設計，該方法會先分析影像的像素值（或誤差值）的分佈情形，並且將機密訊息嵌入頻繁出現的峰值。然而，直方圖位移技術的藏入能力會受到統計後的峰值個數影響，且影像品質將取決於峰值和從未出現的零值之間的距離所影響，假使峰值和零值的距離越遠，或是介於兩者之間的像素越多，會將導致影像嚴重失真。因此，本研究採用邊緣敏感度分析的Lukac預測器以及Fang和Fan的預測器進行設計改良。在第一篇研究中，利用標準差判斷區塊的複雜程度，並且設計兩個自適性調整的門檻值，用以區分區塊是平滑還是複雜，若是平滑區塊，則使用適合Lukac預測器的平滑區塊，反之，則以及適合使用Fang和Fan預測器的複雜區塊；。在第二篇研究中，我們著重於影像被多次進行嵌入時，能夠使像素進行還原並且降低影像失真程度。故延申用第一篇研究的兩種預測器，用以提升預測精確度和藏入能力，並且採用Chen等學者的像素互補機制進行直方圖位移，藉此減少影像失真度。

　　近期可逆式資訊隱藏技術發展出雙影像嵌入策略，該方法先將影像複製成兩張相同的影像，並且利用像素修改規則表嵌入機密訊息和像素修改。本論文提出的第三篇研究利用最低顯著位元匹配法進行嵌入，並且針對無法還原的七個規則設計不同的像素修改方式，藉此提升整體影像品質和藏入能力。

關鍵字

像素修改規則表 ； 雙影像技術 ； 不對稱直方圖位移 ； 標準差 ； 邊緣敏感度分析 ； 可逆式資訊隱藏 ； 直方圖位移技術 ； 最低顯著位元匹配法

並列摘要

Histogram shifting is a common reversible data hiding method that can effectively embed secret information using the distribution of image pixel values (or errors). However, the hidden capacity of image depends on the numbers of peak point, and the image quality of histogram shifting depends on the distance between the zero point and peak point, and a long distance can cause serious image distortion. Therefore, we propose two methods on an improved edge-sensing analysis technique utilizing combinations of Lukac’s method and Feng and Fan’s method. The first method uses standard deviation calculations to analyze complex images, and determines which prediction method can be applied based on standard deviation thresholds. The second method employs integration of the asymmetric-histogram shifting (established by Chen et al.) in order to restore the error value to a place near the original image pixel value in the second shift. The experimental results show that a high image quality is achieved. Images with different characteristics will have better information capacity and image quality using the two predictive methods.

　　In recent years, the dual imaging technique is used often in reversible data hiding method. It creates two copies of the original image for embedding a message to achieve high hiding capacity. Image quality and hiding capacity in the dual imaging method are affected by embedding (or pixel modification) rules. This study proposes an alternative approach. It is based on least-significant-bit matching (LSB-matching) method for embedding using the dual imaging technique. It employs seven rules to camouflage pixel modifications. According to experimental results, we found that the method proposed in this study maintains a relatively high camouflage image quality with a high hiding capacity. The quality of the two camouflage images is of an above average level. In addition, a better image quality is also obtained from natural images with various characteristics.

並列關鍵字

Dual imaging technique ； Asymmetric-histogram shifting ； Standard deviation ； Edge-sensing analysis ； Reversible data hiding ； Histogram shifting ； Embedding rule ； Least-significant-bit matching

參考文獻

[45] J. Wen, J. Lei, and Y. Wan, “Reversible Data Hiding Through Adaptive Prediction and Prediction Error Histogram Modification,” International Journal of Fuzzy Systems, Vol. 14, No. 2, pp. 244-256, 2012.

[1] A.M. Alattar, “Reversible Watermark Using the Difference Expansion of a Generalized Integer Transform,” IEEE Transactions on Image Processing, Vol. 13, pp. 1147-1156, 2004.

[2] O.M. Al-Qershi and B.E. Khoo, “Two-Dimensional Difference Expansion (2D-De) Scheme with a Characteristics-based Threshold,” Signal Processing, Vol. 93, pp. 154-162, 2013.

[4] I.C. Chang, Y.C. Hu, W.L. Chen, and C.C. Lo, “High Capacity Reversible Data Hiding Scheme Based on Residual Histogram Shifting for Block Truncation Coding,” Signal Processing, Vol. 108, pp. 376-388, 2015.

[5] C.C. Chang, T.D. Kieu, and Y.C. Chou, “Reversible Data Hiding Scheme Using Two Steganographic Images,” TENCON 2007 - 2007 IEEE Region 10 Conference, pp. 1-4, 2007.