VIP内容

在当前大规模数据检索任务中,学习型哈希方法能够学习紧凑的二进制编码,在节省存储空间的同时能快速地计算海明空间内的相似度,因此近似最近邻检索常使用哈希的方式来完善快速最近邻检索机制。对于目前大多数哈希方法都采用离线学习模型进行批处理训练,在大规模流数据的环境下无法适应可能出现的数据变化而使得检索效率降低的问题,提出在线哈希方法并学习适应性的哈希函数,从而在输入数据的过程中连续学习,并且能实时地应用于相似性检索。首先,阐释了学习型哈希的基本原理和实现在线哈希的内在要求;接着,从在线条件下流数据的读取模式、学习模式以及模型更新模式等角度介绍在线哈希不同的学习方式;而后,将在线学习算法分为六类:基于主-被动算法、基于矩阵分解技术、基于无监督聚类、基于相似性监督、基于互信息度量和基于码本监督,并且分析这些算法的优缺点及特点;最后,总结和讨论了在线哈希的发展方向。

http://www.joca.cn/CN/abstract/abstract24489.shtml

成为VIP会员查看完整内容
0
11

最新论文

Due to the widespread of advanced digital imaging devices, forgery of digital images became more serious attack patterns. In this attack scenario, the attacker tries to manipulate the digital image to conceal some meaningful information of the genuine image for malicious purposes. This leads to increase security interest about protecting images against integrity tampers. This paper proposes a novel technique for protecting colored images against forgery and pixel tamper. The proposed approach is designed as a hybrid model from three security techniques, Message Digest hashing algorithm (MD5), Advanced Encryption Standard-128 bits (AES), and Stenography. The proposed approach has been evaluated using set of image quality metrics for testing the impact of embedding the protection code on image quality. The evaluation results proved that protecting image based on Least Significant Bit (LSB) is the best technique that keep image quality compared with other two bit-substitution methods. Moreover, the results proved the superiority of the proposed approach compared with other technique in the literature.

0
0
下载
预览
父主题
Top