Abstract: Now days, Cloud computing itis been playing a crucial role in terms of data storing and reducing the overallcost to entrepreneurs. Storage requirements for visual data have increased inrecent years, following the appearance of many interactive multimedia servicesand applications for mobile devices in personal and business scenarios. Thiswas a key a determining factor for the adoption of cloud-based data outsourcingsolutions. However, even the outsourcing of data storage in the cloud leads tonew security challenges that must be carefully addressed. We propose a secureframework for the storage and recovery of the subcontracted privacy protectionin large archives of shared images. Our proposal is based on IES-CBIR, a novelEncryption scheme of the image that presents image recovery properties based oncontent. The framework allows both encrypted storage and search usingcontent-based image retrieval queries while preserving privacy against honestbut curious cloud administrators. We have built a prototype of the proposedframework, formally analyzed and tested its safety properties, andexperimentally assessed its performance and accuracy of recovery. Our resultsshow that IES-CBIR is probably safe, allowing more efficient operations thatthe existing proposals, both in terms of complexity of time and space, andopens the way to new scenarios of practical application.
Keywords: Cloud computing, Data andcomputation outsourcing, encrypted data processing, searchable encryptioncontent based image retrieval.