The Study of Reversible Data Hiding Based on Histogram Shifting and Adaptive Group Pixel Differencing

碩士 === 銘傳大學 === 電子工程學系碩士班 === 103 === There are already a lot of data hiding methods. Steganography of data hiding in spatial domain can be divided into reversible and non-reversible approaches. The capacity of non-reversible data hiding methods is usually more than that of the reversible techniques...

Full description

Bibliographic Details
Main Authors: Yen-Chen Chen, 陳延承
Other Authors: Ping-Sheng Huang
Format: Others
Language:zh-TW
Published: 2015
Online Access:http://ndltd.ncl.edu.tw/handle/83263356921920133799
Description
Summary:碩士 === 銘傳大學 === 電子工程學系碩士班 === 103 === There are already a lot of data hiding methods. Steganography of data hiding in spatial domain can be divided into reversible and non-reversible approaches. The capacity of non-reversible data hiding methods is usually more than that of the reversible techniques. However, image information are normally sacrificed at the same time. On the other hand, reversible data hiding approaches can recover the stego-image to the original image after data extraction. This thesis presents a high-capacity reversible data hiding algorithm, based on histogram shifting and adaptive group pixel differencing. The algorithm can not only use the information of adjacent pixel difference values, the histogram peak points can also be made more concentrated. Furthermore, the property of n×n group pixels and double peak points are adopted to increase the capacity of data hiding. Finally, the method of double-bit data hiding is proposed and the histogram shifting is not limited only at the peak point. Using the method of double peak points in data hiding, the capacity can reach to an average of 0.2 bpp (bit per pixel) and an average PSNR value of 45dB can be accomplished. Furthermore, by combining the method of double peak points and double-bit data hiding, the capacity can be enhanced to an average of 0.3 bpp and an average PSNR value of 40dB can be maintained.