可逆式資訊隱藏技術應用雙影像及邊緣敏感度預測方法

Reversible Data Hiding Techniques Based on Dual Images and Image Quality Edge Sensitive Prediction

許永明(Yung-ming Hsu)

指導教授 ： 呂慈純

朝陽科技大學/資訊學院/資訊管理系/碩士(2014年)

摘要

近年來，許多專家學者提出可逆式資訊隱藏技術來保護資料，如：Lee和Huang學者於2011年提出了一個雙影像可逆式資訊隱藏技術，該技術進行藏入動作時，僅會對像素進行加減1之修改，故可得到一張高品質的偽裝影像，然而，此方法每四個像素僅能藏入五個機密位元，藏入量仍有改善的空間，因此本研究提出一個高藏入效益的雙影像技術方法，所提方法對每個像素可藏入T個機密位元，其中T為一個自適應的門檻值，故能藉由T控制藏入量和偽裝影像品質。實驗結果顯示所提方法除了達到T 位元的藏入效益外，所產生的偽裝影像品質也優於Lee和Huang的方法。
雙影像藏入法技術雖可保持非常不錯的影像品質，但收送方皆須保留二張影像，對收、送方不免造成一些困擾，因此，本研究另外提出一個單影像可逆式資訊隱藏技術，該技術先利用Lukac預測法以產生預測值，再透過所提出的藏入規則將機密資訊嵌入於預測誤差值中。由於該規則僅針對預測誤差值進行加減1之修改，故偽裝像素與原始像素之間差異是相當小的。因此，若重複藏入機密資訊時，使用這些偽裝像素所產生的預測值與使用原始像素所產生的預測值會非常相似，故可允許嵌入更多機密資訊，並且維持良好的影像品質，實驗結果顯示所提方法可達到高藏入率與保持高影像品質。

關鍵字

Lukac預測法 ； 可逆式資訊隱藏技術 ； 自適應門檻值 ； 藏入量 ； 邊緣敏感度預測 ； 影像品質 ； 偽裝影像 ； 雙影像

並列摘要

In recent years, many scholars have proposes reversible information hiding methods to protect confidential data. For example, Lee and Huang propose a dual-image reversible information hiding method in 2011. They modified the pixels by adding or subtracting one to embed secret data. Consequently, the quality of stego images obtained by the above modification method is excellent. However, in their method, each set of four pixels can be used to embed five secret bits, therefore restricting the hiding capacity. In this thesis, we propose a reversible information hiding method with high embedding rate to improve the drawback of Lee and Huang’s method. In the propose method, each pixel can be used to embed T secret bits, where T is an adaptive threshold that can control the image quality and hiding capacity. Experimental results show that the embedding rate of the propose method achieves T bpp, which is higher than that of the recently propose methods. Our method can embed secret data and remain good image quality. However, the technique generates two stego images that may cause some inconvenience for the sender and the receiver.
Hence, this thesis proposes another single-image reversible data hiding technique that uses Lukac prediction method to produce a prediction value and then embeds the secret data into the prediction error by using the propose embedding rules. Since the propose rules only increase or decrease the prediction error by one unit, the difference between the stego pixel and original pixel is small. Thus, for the iterative embedding, the prediction value generated by the stego pixel is very similar to that generated by the cover pixel. As a result, the pixel allows for embedding of more secret data without huge image distortion. Experimental results show that the proposed method can accomplish high hiding rate and preserve high image quality.

並列關鍵字

edge sensitive prediction. ； Lukac prediction method ； reversible data hiding technique ； image quality ； adaptive threshold ； hiding capacity ； stego image ； dual image

參考文獻

[2] Chan, C. K. and Cheng, L. M., “Hiding Data in Images by Simple LSB Substitution,” Pattern Recognition, Vol. 37, No. 3, pp. 469-474, 2004.

[3] Kieu, T. D. and Chang, C. C., “A Steganographic Scheme by Fully Exploiting Modification Directions,” Expert Systems with Applications, Vol. 38, pp. 10648-10657, 2011.

[4] Lin, C. C., “An Information Hiding Scheme with Minimal Image Distortion,” Computer Standards & Interfaces, Vol. 33, No. 5, pp. 477-484, 2011

[5] Lee, C. F., Chen, H. L. and Tso, H. K., “Embedding Capacity Raising in Reversible Data Hiding Based on Prediction of Difference Expansion,” Journal of Systems and Software, Vol. 83, No. 10, pp. 1864-1872, 2010.

[6] Lee, C. F. and Huang, Y. L., “Reversible Data Hiding Scheme Based on Dual Stegano-images Using Orientation Combinations,” Telecommunication Systems, pp. 1-11, 2011.