With the rapid growth of the number of images, many content-based image retrieval methods have been extensively used in our daily life. In general, image retrieval services are very expensive in terms of computing and storage. Therefore, outsourcing services to the cloud server is a good choice for image owners. However, privacy protection can become a big issue for image owners because the cloud server can only be semi-trusted. In this paper, we propose a novel image retrieval scheme. It is a ciphertext image retrieval method based on random mapping features with the bag-of-words model. After encrypting the image with Advanced Encryption Standard and block permutation, the cloud server generates random templates and then extracts the local features. All local features are clustered by k-means algorithm to form the visual word. The histogram of encrypted visual words is constructed in this way as the feature vector to represent each image. The similarity between images can be measured by the distance between feature vectors on the cloud server. Experiments and analysis prove the effect of the scheme. INDEX TERMS Image retrieval, AES encryption, BOW model, random mapping.
Backup datacenters provide massive data storage and access services, and their failure may result in huge economic losses. So their location selection requires low damage risk and high evacuation capability simultaneously. But previous works on backup datacenter placement have not jointly considered these two factors from the viewpoint of traffic engineering and might result in the unnecessary loss in case of disaster. In this paper, with the global view of network resources in the software defined network scenarios, we propose a new disaster-and-evacuation-aware backup datacenter placement strategy. To reduce backup loss risk and apply rapid post-disaster evacuation, we jointly consider expected disaster loss and evacuation latency and formulate a new disaster-and-evacuation-aware facility location problem (NP-hard) which is multi-objective. To obtain the solution according to the disaster situation assessment, we propose a disasterand-evacuation-aware multi-objective optimization algorithm. We optimize multiple objectives owning different coefficients in different disaster situations. We introduce location-output-capability, backupevacuation-latency, Pareto-recommendation-degree, and node-damage-loss to guide solution searching. We prune the external set according to fitness-deviation-ratio to improve convergence speed and computation efficiency of the algorithm. Through extensive simulations, we demonstrate that our algorithm is efficient and promising with less expected disaster loss and higher evacuation capability simultaneously.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.