一種圖像Hash算法的實現(xiàn)

摘要：隨著圖像處理技術的發(fā)展，人們對數(shù)字圖像內容的更改和編輯變得非常容易，且不留痕跡，用眼睛很難判定其真實性。當數(shù)字圖像中包含有政治、軍事、商業(yè)、監(jiān)控、個人、及醫(yī)療等敏感的或重要的信息時，必須有相關的技術對其進行認證。面向認證的圖像Hash是圖像認證的一種主要的技術手段。近年來受到了廣泛的重視并出現(xiàn)了多種圖象Hash算法。其中Fridrich等人提出的一種魯棒的圖象Hash算法被認為是經典的圖象Hash算法之一。
面向認證的圖像Hash主要包括以下三個性能指標：1) 穩(wěn)健性，2) 脆弱性， 3) 安全性，其是圖像Hash的核心性能指標。為了更深入的研究面向認證的Hash算法的性能，需要實現(xiàn)一些經典的算法。本論文深入地了解Fridrich所提出的算法，并實現(xiàn)了該算法。在實現(xiàn)的過程中重點解決了閾值的選取問題，使圖象Hash值中的0，1個數(shù)盡可能相同。

關鍵詞：圖像認證 圖像安全 密鑰保護 Hash函數(shù)

Image identify based on Hash function
Abstract: With the development of the technology of the processing of image, it is easy for people to change and edit the image and that is invisible to eyes. If the content of the image involved the some critical information such as politician, military, business and medical treatment, We need a kind of technology to identify the picture. Hash function which can identify the image is a major way. There are a lot of Hash functions in recent years. One of them which was offered by Fridrich with robust is widely recognized as a classical arithmetic..
  There are three major indexes of the Hash function : (1)stability (2) frangibility (3)safety. We need to realize some classical arithmetic in order to study the Hash function for image further. This theory analysis the Fridich’s Hash function, and realize the Hash. During the process, how to calculate a threshold so as to the number of the 1 and 0 are approximately equal.