- 會議論文
- OpenAccess
植基於圖像內容與資料隱藏的CAPTCHA-like密碼系統
CAPTCHA-like Content Based Graphical Password by Data Hiding
摘要
文字密碼是目前最常被使用的認證方式。若密碼過長會不容易記，但是太短則密碼強度又不夠。圖像密碼能解決文字密碼的問題，並且可以達到相同的安全性。圖像密碼系統有兩大類：PP(Pass-Points) 圖像密碼與CCP (Cued Click-Points) 圖像密碼。PP 圖像密碼認證系統是在一張影像，依序在圖上點擊數個點。在一張圖上記憶不同點及順序，容易造成順序混淆。CCP 改善了PP 的問題，CCP 是讓使用者在一連串的圖片中，點擊每張圖片中的一點，但它需要儲存大量圖片作為設定密碼使用。若是只讓使用者記憶圖片的順序像CCP一樣，但又不需要點擊每張圖片的點出現不同的圖片，就能解決需要大量圖像的問題。我們採用了植基於內容CAPTCHA-like 的密碼認證觀念。認證是植基於內容，所以圖像若是做旋轉、平移、轉置、縮放、明亮度等簡單改變或從不同角度看都有相同的內容。所以每次認證的圖像內容一樣，但又不是同一張張圖片。我們提出植基於圖像內容與資料隱藏 的CAPTCHA-like 密碼系統(簡稱CCGP)，只需記憶圖形順序就可以通過認證，改善了CCP 需大量圖像儲存空間的問題。但是我們的CCGP 付出的成本是認證伺服器要執行圖像識別的運算。
並列摘要
Most authentication systems use the combination of text password. However, there are two major problems in alphanumeric passwords: the difficulty of remembering (long password makes it difficult to remember), and the security (the strength of short password is not enough). Graphical passwords may tackle this problem of using alphanumeric passwords. There are two types of graphical password systems: Pass-Points (PP) and CCP (Cued Click-Points) graphical passwords. In PP, users should click various points in a single picture. It is not easy to remember different points and their sequence. CCP improves PP, and it allows users to click only one point in a picture. However, CCP needs storing large amounts of pictures in data base. We adopted the CAPTCHA-like password authentication concept. Our authentication is based on the content, and thus rotation, translation, transpose, zoom, brightness, various perspective of pictures have the same content. Therefore, each time the picture has the same content but it is different from previous one. In the proposed CAPTCHA-like content based graphical password (CCGP), users only have to remember a sequence of picture for authentication, so that it can tackle the storage problem of CCP. However, our cost is that the authentication server needs additional operations to perform image recognition.
延伸閱讀
- 鄧弘文（2016）。運用影像辨識於視覺密碼應用系統之設計〔碩士論文，淡江大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0002-0502201613233300
- 林鴻鈞（2004）。新的數位影像加解密演算法及隱藏式視覺密碼演算法〔碩士論文，大同大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0081-0607200917233649
- 黃培修（2007）。結合非擴展式視覺密碼與機率理論於數位影像之保護〔碩士論文，淡江大學〕。華藝線上圖書館。https://doi.org/10.6846/TKU.2007.00090
- LIU, X., ZHANG, Y., HU, J., TANG, M., & GAO, H. (2019). Breaking Text-Based CAPTCHAs using Average Vertical Partition. Journal of Information Science and Engineering, 35(3), 611-634. https://doi.org/10.6688/JISE.201905_35(3).0008
- Zhuang, X., Chang, C. C., Wang, Z. H., & Zhu, Y. (2014). A Simple Password Authentication Scheme Based on Geometric Hashing Function. International Journal of Network Security, 16(4), 271-277. https://doi.org/10.6633/IJNS.201407.16(4).05