一种云环境下图像的安全检索方法

摘要

In order to protect data privacy,the image with sensitive or private information needs to be encrypted before being uploaded to cloud server.However,this causes difficulties in image retrieval.In this paper we propose a secure image retrieval method under the cloud environment.It carries out intra-block scrambling and inter-block scrambling on DCT (discrete cosine transform)coefficients of R and G channels in image colour space separately,and makes intra-block scrambling encryption on B channel;and then it extracts the colour moment,the entropy of information and the inter-block LBP (local binary patterns)as image’s feature vectors,and determines the similarity of images through comparing Manhattan distance between these feature vectors.The proposed method ensures both the image confidentiality and retrieval accuracy,and is able to achieve secure image retrieval.%为了保护数据机密性，具有敏感信息的图像在上传到云服务器之前需要进行加密，然而，这样做会给图像检索带来问题。提出一种云环境下图像的安全检索方法，对图像颜色空间的 R 和 G 通道的 DCT（Discrete Cosine Transform）系数分别进行块内置乱和块间置乱，对 B 通道进行块内置乱加密；然后提取颜色矩、信息熵以及块间 LBP（Local Binary Patterns）作为图像的特征向量，通过比较图像的特征向量之间的曼哈顿距离来确定图像的相似度。该方法保证了图像的机密性和检索精确度，能实现对图像的安全检索。