Please use this identifier to cite or link to this item:
標題: (k, n)影像之可逆式資訊隱藏技術
(k, n)-Image Reversible Data Hiding
作者: Ying-Hsuan Huang
關鍵字: (k, n)-image reversible data hiding;difference expansion;prediction error expansion;histogram shifting;dual stego images;hiding capacity;data encoding;PSNR value;可逆式資訊隱藏;差值擴張;直方圖位移;雙偽裝影像;藏入量;(k, n)影像之可逆式資訊隱藏;機密資訊編碼;影像品質
In the past years, information security techniques have been proposed extensively, e.g., cryptography, secret sharing, data hiding, and so on. Different techniques have their own advantages and disadvantages. For example, the secret data can be encrypted by cryptography to be a meaningless ciphertext. The hackers cannot effectively derive from the ciphertext. However, they can intercept and destroy the ciphertext.
Different from cryptography, data hiding method is to embed secret data into one cover image to generate one stego image. The hackers cannot effectively discriminate between the stego image and the cover image, thereby avoiding hacker's doubt and attack. Reversible data hiding methods can be classified into five types, i.e., difference expansion, prediction error expansion, histogram shifting, dual stego images and reversible contrast mapping. These methods can extract the embedded secret data and recover the original image. However, the hiding capacity of these methods can be enhanced further.
In order to embed more secret data, this dissertation presents a (k, n)-image reversible data hiding method, where k denotes the number of secret images, n denotes the number of stego images and k ≤ n. The embedding capacity becomes greater as n increases. In addition, the proposed method can embed (n – k) cheating images to cheat hackers or increase their analysis time. Even if some stego images are lost, the cover image can be recovered effectively and the secret data are extracted correctly.
Under the environment with less computational cost, the proposed method only embeds secret data into one cover image by expanded the difference between pixels. However, embedding the secret bit '1' into one cover pixel may cause serious distortion of the image. In order to maintain good quality of the stego image, this dissertation proposed a secret data encoding strategy that can effectively transform secret bits from '1' into '0.' Embedding the encoded data into the cover image does not invoke serious distortion. Experimental results showed that the proposed method has greater embedding capacity and better image quality than previous methods.
其他識別: U0005-2107201514565800
Rights: 同意授權瀏覽/列印電子全文服務,2018-07-27起公開。
Appears in Collections:資訊科學與工程學系所

Files in This Item:
File Description SizeFormat Existing users please Login
nchu-104-8098056007-1.pdf4.92 MBAdobe PDFThis file is only available in the university internal network    Request a copy
Show full item record

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.