Research On Encrypted Image Retrieval Method Based On Local Binary Pattern

The rapid growth of digital images motivates organizations and individuals to outsource image storage and computation to the cloud.However,the defenseless upload will raise the risk of privacy leakage.And the existing encrypted image retrieval technology makes users undertake a lot of computing tasks.Therefore,this paper proposes an encrypted image retrieval scheme using LBP(Local Binary Pattern)feature.The BOW(Bag of Words)model and deep learning technology are utilized to extract feature from encrypted image,and feature extraction and index construction are outsourced to the cloud server to reduce the computational burden of users.The main research of this paper is as follows:A privacy-preserving image retrieval scheme using big-block permutation,3×3block permutation within big-blocks,pixel permutation within 3×3 blocks,and polyalphabetic cipher.The use of polyalphabetic cipher improves security and causes no degradation in terms of retrieval accuracy as the substitution tables are generated by the order-preserving encryption.In this way,secure Local Binary Pattern(LBP)features can be directly extracted as the local features from the encrypted big-blocks,which is efficient as there is no communication between the cloud server and image owners to do so.The secure local LBP features are utilized to generate image feature for every image by the Bag-of-Words(BOW)model.A privacy-preserving image retrieval scheme combining deep learning technology and LBP features is proposed.The block permutation and intra-block pixel substitution with multiple order-preserving tables were used to protect images.The encrypted images would be then uploaded to the cloud server and the LBP maps of encrypted images were calculated on the cloud side.The neural network was trained by the encrypted LBP maps and the image features were extracted from them with the trained neural network.Image index was established by the extracted features.The security analysis and experimental result indicate that this scheme is better than the many previous schemes in terms of security and retrieval accuracy.