Research on Design of New Reversible Data Hiding Scheme

碩士 === 國立虎尾科技大學 === 光電與材料科技研究所 === 98 === In 2006, Ni et al. proposed a data hiding method based on histogram. Its characteristic is to shift the straights between the peak and the zero point in histogram. Then, it hides the information into the peak and adjacent pixel value. It can not only insert...

Full description

Bibliographic Details
Main Authors: Yan-Hung Lin, 林彥宏
Other Authors: Wen-Chung Kuo
Format: Others
Language:zh-TW
Online Access:http://ndltd.ncl.edu.tw/handle/jxr8bd
Description
Summary:碩士 === 國立虎尾科技大學 === 光電與材料科技研究所 === 98 === In 2006, Ni et al. proposed a data hiding method based on histogram. Its characteristic is to shift the straights between the peak and the zero point in histogram. Then, it hides the information into the peak and adjacent pixel value. It can not only insert into high-capacity data, but also keep excellent quality of visual. But the sender must send the original the highest point and lowest point of information to the receiver, the receiver can correctly extract the confidential information and restore the carrier image. Hwang et al. proposed the method that unchanged the peak in the histogram in the same year. They improved Ni et al.’s drawback, furthermore enhanced the capacity. In 2007, Fallahpour and Sedaaghi divided image into several blocks. They hide the security into those blocks. Theirs increase a great deal of capacity, but make more additional information. Since Ni et al. proposed the idea of histogram, follow-up researchers focus on the enhancement of capacity. We discover some characteristics about data hiding based on histogram. We propose a new high-capacity method based on those characteristics and image dividing. By the concept of image dividing and characteristics based on histogram, our method needs not to transmit any additional information to extract the secret. At one time, it restores the stega-image to the original image. On rectangle images, we can use different dividing to change characteristics based on histogram. The different dividing will reduce probability which can be detected.