在數位博物館中利用浮水印、資訊隱藏與影像驗證技術作視覺性資料版權與註解保護

Abstract

典藏於博物館中的文物數位化之後，其相關版權與註解性資料必須被保護。本論文針對三種影像格式，首先提出了隱藏註解性資料的方法來簡化影像與其註解性資料之關聯。為了在BMP圖檔中隱藏註解性資料，我們在本論文中提出了一種新的方法來隱藏可以用來定位註解性資料起點的定位線訊號，並提出了一種新的方法來隱藏註解資料或影像於含有調色盤的GIF圖檔中，我們並提出了一種利用人眼無法察覺的方式來隱藏浮水印訊號於三種不同圖檔格式中的方法。為了在GIF圖檔中隱藏如標誌等類的二位元資料，我們亦提出了一種可行的影像浮水印方法，最後，我們亦提出了一種在三種圖檔格式中隱藏易碎浮水印訊號的方法，來驗證影像的完整性與真實性。為了驗證BMP影像，我們提出了一種新的抗切割易碎浮水印方法，並提出了一種根據排序調色盤來隱藏易碎浮水印的方法以用來偵測並定位竄改，最後提出實驗結果來證明所提方法的可行性。

After artworks preserved in museums are digitized, the copyright and annotation data of them need to be protected. Proposed in this study are methods for embedding annotations within images of three types of formats to facilitate the association between images and their commentary data. To embed annotation data within BMP images, a novel method for embedding boundary line signals, which can be used for localizing starting points of annotations within stego-images, is proposed in this study. A new data hiding method is also proposed for embedding annotation data or image data within GIF images whose formats include color palettes. Also proposed are methods for embedding watermark signals imperceptibly within the three types of images to prove its copyright owner. To embed binary data like logos in GIF images, a practical method for image watermarking is proposed. Finally, methods for embedding fragile signals within the three types of images for image authentication to verify image integrity and fidelity is also proposed. For verification of a BMP image, a new anti-cropping fragile watermarking method is proposed. Another fragile watermarking method based on a sorted palette for GIF images is also proposed for detecting and localizing image tampering. Good experimental results prove the feasibility of the proposed methods.