基於人類視覺系統之可適應性三維模型資訊偽裝技術

ASIA unversity > 資訊學院 > 行動商務與多媒體應用學系 > 科技部研究計畫 > Item 310904400/9599

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: http://asiair.asia.edu.tw/ir/handle/310904400/9599

题名:	基於人類視覺系統之可適應性三維模型資訊偽裝技術
作者:	蔡淵裕
贡献者:	資訊學院資訊多媒體應用學系
关键词:	三維模型資訊隱藏可適應性人類視覺系統 3D Models Data Hiding Adaptability Human Visual System
日期:	2009
上传时间:	2010-05-13 02:43:08 (UTC+0)
摘要:	隨著個人電腦的普及和資訊科技的蓬勃發展，人們對網際網路之依賴日漸提高，網路服務在日常辦公和生活中，儼然已位居異常重要的地位。然而，人們使用網路時，常因缺乏適當的規範和警覺，造成網路遭到濫用，甚至出現惡意使用網路的犯罪行為，侵害其他網路使用者的基本權益，為了保護智慧財產權以及維護資訊安全，資訊隱藏(Information Hiding)演算法之研究乃因應而起。先前大多數的資訊隱藏主要是將資訊嵌入在常見的數位媒體中，例如：靜態影像、動態影像、音訊等。然而，近年來三維電腦遊戲動畫與三維掃瞄技術等相關研究，使得三維模型被廣泛地使用，因此三維幾何模型的資訊隱藏演算法之研究便隨之產生。現今之三維模型資訊偽裝技術，強調之重點大多在於高隱藏量、低失真度、正確擷取秘密訊息等部分，若能同時支援多邊形模型與點模型則更為理想。然而，此類的演算法通常有一缺失，即是在每個頂點上均嵌入相同的資料量。相同資料量對於不同區域的頂點，所造成的影響也會不同，因此會造成很明顯的落差而引人注意，違背了資訊隱藏之目的。為了解決上述的問題，在本計畫中，我們提出一個三維模型資訊偽裝演算法，除了保有現今偽裝技術之基本功能外，同時整合三維模型幾何特性以及人類視覺靈敏度，發展出一套新式且可適應性之三維模型資訊偽裝技術，針對擁有不同表面特性的頂點嵌入不同數量的秘密訊息，將在此技術領域中，提供突破性的貢獻。 As the popularity of personal computers and the flourishing development of information technologies, people have created more and more dependence on the internet. The internet service has played an important role in everyday life. Being lack of proper standards and awareness, however, the internet services are usually misused and even bring cyber crimes to infringe upon the lawful rights of others. In order to protect the intellectual property rights and maintain the information security, the research on information hiding algorithms have sprung up. In prior years, many steganographic and watermarking algorithms have been researched. However, they often use digital multimedia data, such as still image, moving image, audio and video, as cover media to embed the secret message. Recently, with the development of various 3D applications, such as in the computer animation and 3D laser scanning techniques, the use of three-dimensional models in information hiding techniques has gradually attracted attention. The emphasis for current 3D steganographic algorithms are mostly on high data capacity, low distortion and correct data extraction. It would be much better to support both polygonal models and point geometries. However, the disadvantage of them is the same data capacity for each vertex in the 3D models. Embedding the same capacity in the vertex located on the surface with different properties may cause the obvious distortion and thus obey the initial goal of the information hiding techniques. To remedy the above problem, in this project, we propose a 3D steganographic algorithm, which integrates the key points of current techniques, the geometric characteristics and human visual system to develop a new and adaptive one. According to the different surface properties that each vertex is located on, the different amount of secret message is embedded. Thus, the important shape features will be preserved and the visual distortion between cover and stego model will be much smaller. We believe that steganography on 3D models with adaptability will do a great contribution to the area of 3D information hiding.
显示于类别:	[行動商務與多媒體應用學系] 科技部研究計畫

文件中的档案:

档案	描述	大小	格式	浏览次数
98蔡淵裕1.pdf		68Kb	Adobe PDF	573	检视/开启
98蔡淵裕2.pdf		37Kb	Adobe PDF	424	检视/开启

在ASIAIR中所有的数据项都受到原著作权保护.

数据加载中.....