基于哈希算法的图像拷贝检测

在局部保持投影算法（LPP）及最大熵模型的基础上提出了一种基于哈希算法的图像拷贝检测算法。该算法通过对LPP进行优化,可以对已有的图像特征进行快速降维处理,并避免过拟合现象的出现。同时,将低维度特征映射为紧凑的二进制哈希码,大大减少了存储和计算开销,能在千万级图像数据库上实现高效拷贝检测。通过实验确定了使拷贝检测质量最优的哈希码长度和汉明码。实验结果表明,该算法能够获得很好的拷贝检测效果。     

基于散列的长链处理算法

针对铁路信号产品无线闭塞中心、临时限速服务器、列控中心软件对于长链的处理存在的差异进行分析,提出对于长链的一种散列处理算法,兼容目前使用的两种长链规范标准,力求配置最少的信息,得到对于长链计算所需的各种关键参数,以提供给研发、测试开发所需对于长链处理的公共处理接口,使开发人员不再处理长链的计算,解决数据配置人员由于各产...     

DYNAMIC REQUEST DISPATCHING ALGORITHM FOR WEB SERVER CLUSTER

Distributed architectures support increased load on popular web sites by dispatching client requests transparently among multiple servers in a cluster. Packet Single-Rewriting technology and client address hashing algorithm in ONE-IP technology which can ensure application-session-keep have been analyzed, an improved request dispatching algorithm which is simple, effective and supports dynamic load balance has been proposed. In this algorithm, dispatcher evaluates which server node will process request by applying a hash function to the client IP address and comparing the result with its assigned identifier subset; it adjusts the size of the subset according to the performance and current load of each server, so as to utilize all servers＇ resource effectively. Simulation shows that the improved algorithm has better performance than the original one.     

视频哈希的安全性分析

为了对鲁棒多媒体哈希的安全性进行定量分析,用Shannon的唯一解距离来解释鲁棒多媒体哈希的安全性.以一个典型的鲁棒视频哈希算法为例,对该算法的特征提取方式进行数学建模,并针对不同的攻击类型,用Shannon的唯一解距离测量该鲁棒视频哈希算法的安全级别,定量分析其安全性.分析结果表明:鲁棒视频哈希算法一个密钥的重复使用次数不能超过所求得的唯一解距离,否则,算法的密钥可以被估计出来.     

Image Betrayal Checking Based on the Digital Fingerprint of Images on the Internet

In the Internet environment, documents are easily leaked, and divulged files spread rapidly. Therefore, it is important for privacy institutions to actively check the documents on the Internet to find out whether some private files have been leaked. In this paper, we put forward a scheme for active image betrayal checking on the Internet based on the digital fingerprint, which embeds fingerprints into privacy documents, extracts codes from the Internet images, and then finds out the divulged files by matching two groups of codes. Due to so many documents on the Internet, the number of times of code comparison is huge, which leads to a large running time. To overcome the deficiency in practical application, we optimized the process by accurate matching methods and approximate matching method. Then a method was proposed to group objects by locality sensitive hashing (LSH) process before code comparison, in order to eliminate the vast majority of unrelated pairs. Experiments prove that this method could operate with less running time and less memory.