基于3D-DWT及混沌映射的加密医用体数据鲁棒检索算法研究

摘要

随着医疗数据的共享日益频繁,数据安全问题变得日益突出.安全起见,把医学图像加密后存储到服务器或云端.如何实现对加密医学图像检索成为一个亟待解决的问题.提出一种基于混沌和三维小波变换的加密医用体数据鲁棒检索算法.其主要步骤为:首先,通过三维小波变换和混沌映射对医用体数据进行加密;然后,对加密体数据进行小波分解,得到子带图像,再对子带图像进行DCT变换,提取DCT变换能量聚集的低频系数作为加密体数据的特征向量,建立加密图像的特征向量数据库;最后,通过计算加密图像的特征向量的相似度来检索目标图像,将检索到的加密图像返回,并解密所检索到的医用体数据.实验结果表明,该检索算法对加密体数据的常规攻击和几何攻击具有较强的鲁棒性,并提高了图像检索的安全性和准确性.%With the increasing sharing of medical data,data security issues become increasingly prominent.For security reasons,we encrypt medical image before storing them in the server or cloud.How to realize the retrieval of encrypted medical image becomes an urgent problem to be solved.This paper proposes an encrypted medical volume data robust retrieval algorithm based on chaos and three-dimensional wavelet transform.The main steps are as follows:first,medical data volume is encrypted by three-dimensional wavelet transform and chaotic mapping;then,the encrypted volume data is decomposed by wavelet to obtain the subband image,and then perform DCT transform on the sub-band image,the low frequency coefficients of DCT transform energy aggregation are extracted as feature vector of the medical data,and the features vector database of encrypted image is established;finally,the target image is retrieved by calculating the feature vector similarity of encrypted image,the retrieved encrypted image is returned,and decrypt the retrieved medical volume data.The experimental results show that the proposed retrieval algorithm has strong robustness against common attacks and geometric attacks of encrypted volume data and improves the security and accuracy of image retrieval.