以預測器為基礎之資訊隱藏技術

Predictor-based Data Hiding Techniques

陳昌睦(Chang-Mu Chen)

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

隨著科技的進步，近年來已經有許多資訊隱藏技術被提出，這些技術依是否可恢復原始影像分為兩類，第一類為可逆式資訊隱藏，第二類為不可逆資訊隱藏，本研究針對可逆式資訊隱藏與不可逆式資訊隱藏分別提出不同方法。
不可逆式資訊隱藏方法部分是改良Hsiao與Chang學者於2011年提出一個利用恰可察覺失真度(Just Noticeable Distortion, JND)為基礎的自適應資訊隱藏方法，該方法利用JND模型判斷是否可嵌入機密訊息，然而JND值的計算非常耗時且複雜，因此本研究改良Hsiao與Chang學者提出方法，以不使用JND值進行藏入，利用預測器及透過2的降冪次方與2的補數進行藏入，實驗結果證實本研究不需衡量可失真值，即能達到相似或更佳的影像品質與藏入量。
可逆式資訊隱藏部分是以Chen等學者於2012年提出不對稱直方圖方法為概念，透過GAP梯度預測(Gradient Adjusted Prediction, GAP)、MED邊緣預測(Median Edge Detect, MED)及INP(Interpolation by Neighboring Pixel, INP)預測值產生方式，以預測誤差直方圖修改將機密訊息嵌入至影像中，本論文將分析以不同預測器產生預測誤差直方圖對於嵌入量與影像品質的影響。

關鍵字

梯度預測 ； 二進制補數 ； 恰可察覺失真度 ； 資訊隱藏 ； 邊緣預測

並列摘要

In recent years, many data hiding techniques have been proposed, which can be classified into two types according to the reversibility of image, i.e., reversible data hiding and irreversible data hiding. In 2011, Hsiao and Chang proposed an irreversible data hiding method. This method utilizes just noticeable distortion (JND) techniques to determine whether a pixel can be included in hidden information based on its value, but the calculation procedure is very complex. Therefore, this thesis improves Hsiao and Chang’s method through the use of an alternate technique to replace JND techniques, specifically through the use of the 2's complement to embed hidden information in an image. Our experimental results demonstrate that our approach achieves high quality and hidden capacity without the need for JND techniques and their associated computational cost. In 2012, Chen proposed an asymmetric-histogram reversible data hiding method, the hiding capacity of which is satisfactory. However, the predictor in the method is rough, which led to worse prediction efficiency. Consequently, this thesis combines the asymmetric-histogram and different predictors, e.g., gradient adjusted prediction (GAP), median edge detect (MED) and interpolation by neighboring pixel (INP), to achieve a higher hiding capacity and better image quality.

並列關鍵字

Gradient Adjusted Prediction ； Just Noticeable Distortion ； 2's Complement ； Data Hiding ； Median Edge Detect

參考文獻

[11] L. C. Huang, L. Y. Tseng and M. S. Hwang(2012), “The Study on Data Hiding in Medical Images,” International Journal of Network Security, Vol.14, No.6, pp.301-309.

[3] L. An, X. Gao, Y. Yuan and D. Tao(2012), “Robust Lossless Data Hiding using Clustering and Statistical Quantity Histogram,” Neurocomputing, Vol. 77, No. 1, pp. 1-11.

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

[5] C. H. Chou and Y. C. Li(1995), “A Perceptually Tuned Subband Image Coder based on the Measure of Just-Noticeable-Distortion Profile,” IEEE Transactions on Circuits and Systems for Video Technology, Vol. 5, No. 6, pp. 467-476.

[6] 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.