The research on digital watermarking and image recovery

淡江大學機構典藏 > 工學院 > 電機工程學系暨研究所 > 學位論文 > Item 987654321/52501

請使用永久網址來引用或連結此文件: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/52501

題名:	The research on digital watermarking and image recovery
其他題名:	數位浮水印與影像修復之研究
作者:	洪國銘;Hung, Kuo-ming
貢獻者:	淡江大學電機工程學系博士班謝景棠;Hsieh, Ching-tang
關鍵詞:	浮水印;整數餘弦轉換;Zernike 動量;對稱矩陣;影像修補;半色調影像;逆半色調;影像修復;Watermark;IntDCT;Zernike Moment;Symmetric Matrix;inpainting;Halftone Image;Inverse Halftoning;Image Recovery.
日期:	2010
上傳時間:	2010-09-23 17:51:46 (UTC+8)
摘要:	近年來由於通訊科技的方便與快速，使得網際網路成為極佳的數位媒體傳播系統，也因此使得數位媒體非常容易被盜用或竄改，數位數媒體的著作權的保護與與保存成為非常重要的研究課題。數位浮水印的發展可用來保護數位媒體的著作權與完整性，其技術可分成強健型浮水印與易碎型浮水印。強健型浮水印可用來作為數位媒體智慧財產權的保護；易碎型浮水印則可用來作為媒體的認證或被竄改的偵測。在數位浮水印發展的同時，影像修補也是一項非常熱門之研究主題。影像修補的技術，其目的是在修補被破壞影像的區域。影像修補可以延伸到不同的領域，例如：影像編輯、文化遺產的修復、資料的壓縮與網路資料的傳輸等。本篇論文的第一個主軸是介紹一個基於整數離散餘弦轉換的強健型浮水印系統。我們利用4X4的整數離散餘弦轉換，去改善8X8離散餘弦轉換所產生的區塊效應，由於整數離散餘弦轉換的計算只包含了加法與平移運算，且沒有傳統離散餘弦轉換的結尾誤差，因此降低了複雜的計算，並使硬體的實現更加容易。第二個主軸，提出一個基於實對稱矩陣，可以對抗幾何攻擊的半易脆浮水印系統，利用Multi-rings Zernike轉換去偵測影像的旋轉角度，並以實對稱矩陣的主特徵向量所產生的簽章位元來偵測影像受攻擊的區域。本篇論文的第三個主軸，整合了數位浮水印、半色調技術與逆半色調轉換的方法，提出一新的影像修補技術，此技術利用LSB浮水印的技術，嵌入原始影像的半色調影像至原始影像中，用以保護該影像。當影像受破壞時，我們啟動修補程序，以LSB的方法，萃取半色調影像的資訊，並利用LUT逆半色調轉換，得到一參考影像。最後利用此參考影像來完成影像修補的工作。實驗證明，我們的方法在影像修補的效果是非常顯著的。本篇論文最後一個主軸提出一個自動影像認證與修復的方法，以實對稱矩陣的主特徵向量所產生的簽章位元來偵測影像受攻擊的區域後，啟動半色調影像修補機制。我們成功整合我們所提出的易脆浮水印與影像修補技術，建立一個自動的影像恢復系統。 In recent years, Internet is an excellent distribution system for digital media because it is cheap and fast. It is known that digital media can be easily replicated and modified. Thus, digital watermarking has been developed to protect the copyright and integrity of multimedia content. A watermarking scheme can be classified as either robust watermarking or fragile watermarking. Robust watermarking is designed for the copyright and ownership verification. Fragile watermarking is used for image authentication and tamper detection, which can be used to identify any alteration of the images. In parallel to watermarking there was also another field of study, image inpainting. Image inpainting technique aims to fill in the damaged region of an image with new information so that it is hard to find that the image has once been damaged. Image inpainting is extensively used in various fields such as photo editing, cultural relic restoration, data compression and network data transmission. The first issue of this dissertation is to introduce a robust watermarking system based on Integer Discrete Cosine Transform. The 4x4 IntDCT transform can reduce blocking artifacts resulted from 8x8 DCT transform. The integer transform is designed such that the transformation involves only additions and shift operations, and no mismatch exists between the forward and inverse transforms. This reduces the computational complexity and it is much easier for hardware implementation. The second issue of this dissertation is to introduce a new invariant semi-fragile digital watermarking technique based on eigenvalues and eigenvectors of a real symmetric matrix generated by the four pixel-pair. And the multi-rings Zernike transform (MRZT) is proposed to achieve geometric invariance. A signature bit for detecting malicious tampering of an image is generated using the dominant eigenvector. The MRZT method is against the geometric distortions even when the image is under malicious attacks. In the third issue of this dissertation, we combine the algorithms of watermarking, halftoning and inverse halftoning to propose a novel image inpainting technique. This technique use LSB method to embed halftone image into original image for protecting the image. In image repair process, we use LSB method to extract the halftone information, and the reference image is achieved from LUT inverse halftoning. Finally we use the reference image to finish the image inpainting work. Experiment shows the performance of our method is very excellent in image inpainting. In the last issue of this dissertation, a novel scheme of Automatic image authentication and recovery has been proposed, in which the signature bits for detecting malicious tampering of an image are generated using the dominant eigenvector. To achieve high-quality image recovery, we also insert the binary halftone image converted from the original image into the watermarked image to obtain the protected watermarked image. If the image has been tampered, the altered region of the image can be detected and recovered in a fully automated fashion.
顯示於類別:	[電機工程學系暨研究所] 學位論文

文件中的檔案:

檔案	大小	格式	瀏覽次數
index.html	0Kb	HTML	337	檢視/開啟

在機構典藏中所有的資料項目都受到原著作權保護.

TAIR相關文章

資料載入中.....