查詢結果分析
來源資料
頁籤選單縮合
題名 | 資料隱藏偽裝技術之研究=Steganography: An Efficient Data Hiding Method |
---|---|
作 者 | 侯永昌; 趙元甫; 陳培敏; | 書刊名 | 技術學刊 |
卷期 | 15:3 2000.09[民89.09] |
頁次 | 頁363-372 |
分類號 | 312.76 |
關鍵詞 | 資料隱藏; 二元樹; 編碼書; Steganography; Birtree; Codebook; |
語文 | 中文(Chinese) |
中文摘要 | 隨著網路時代的來臨,資訊的流通越來越便利,地球村不再是遙不可及的夢想。藉由網路,人與人之間的連絡更為方便,可以隨時隨地掌握全世界的的資訊。但隨之而來的問題,如資訊倫理的問題,網路安全的問題等等,卻不得不讓我們靜下心來思考,尋求化們的解決方案。 資訊的加解密可以提高資訊在網路上傳送的安全性,但以往我們所使用的加解密方法都有一個共通的缺點,就是所編密出來的密文都是一堆看不懂的亂碼。對於有經驗的網路駭客來說,當他們竊取到這一類的檔案,馬上便會知道這是一個加密過的檔案,於是千方百計的想去破解,就算是再好的一套加密法,也終會有被破解的一天,這只是時間上的問題罷了。因此,近來有人開始研究另一種隱藏資訊的方式,就是把資訊藏在一張看來極為普通的圖片中。由於偽圖是一張普通的影像,所以不容易被懷疑,因此真圖被截取的破壞機會也就小很多了。 本研究是將真圖先利用二元樹來編碼,將相同的色階變成影像區塊,再將每一個影像區堆利用codebook記錄對應到偽圖連續的pixel中。本研究有下列之優點:真圖與偽圖的色階分布並沒有任何限制,不需要色階分布相似。對於真圖的大小並沒有限制,只要編碼後的影像區堆夠少,可以用相當小的偽圖來隱藏真圖,所以小圖帶大圖是可行的。本研究作業方式比較容易,並不需要作色階的平移。而且codebook對應偽圖的pixel也是連續的,因此可以省去search的時間。 |
英文摘要 | The internet make the idea of ‘a global village’ more realistic. People in the world can use it to communicate directly and fast. However some problems occur due to the intensive use of the internet. One of the most critical problems is the security of transmitted data. That is, encryption/decryption of information are serious concerns. There are already many techniques or methods developed to solve this problem. However, the previously developed encryption and decryption techniques have a common disadvantage: the cipher texts after encryption are unrecognizable. Thus, whenever unrecognizable information is retrieved, we know that it may be encrypted and there is some importment information in it. With this knowledge, someone may try to break it and decrypt it. This is just a time issue. Hence researchers started to develop other kinds of methods. One of them is called steganography which is to conceal information in order that the image called the original image, is embedded in another image, called the counterfeit image. Using this method, the snatcher will think the counterfeit image is the data we want to deliver instead of the embedded original image. Even though the snatcher knows about this strategy, he still needs lots of time to uncover it. In this study, we propose an effective data hiding method based on image bintree encoding. That is, in the original image, if the pixels within a block have the same color, they are grouped as a unit corresponding to the pixel with the same color in the counterfeit image. By this way, we can hide a larger original image in a smaller counterfeit image. In this study, we achieve several objectives. The gray value distributions of the embedded and counterfeit images need no be similar. The size of the embedded image can be smaller than that of the counterfeit image. There is no need to perform histogram equalization or vector quantization. A codebook is easy to build and it takes less time to process. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。