Research Of Encrypted Image Searching Technology In Cloud

The advent of the digital age makes surge in the number of images, the image plays an increasingly important role in our life. Store and manage large-scale image database has become an urgent problem to be solved. Thanks to the rapid development of cloud computing technology, users can store their images in the cloud. Cloud storage not only reduces the burden to manage images at the user side, but also provides a convenient data access services for the user. In the cloud storage mode, users’ images are outsourced to the cloud for unified storage and management. However, cloud servers are not entirely credible. The content of users’ images may be exposed to the cloud servers, resulting in leakage of user privacy. Data security and privacy has become the primary problem which restricts the development of cloud computing.In order to protect the image privacy and prevent the image from being illegally accessed, users first encrypt images before outsourcing, and then sent the encrypted images to the cloud. When need to use the image, users first download all the images from the cloud, and decrypt the images for using. However, when the image database reaches a certain size, the usage of "encryption upload, download decrypt" would be very inefficient. This is also contrary to the original intention of convenient use of cloud data. So, efficient use of the encrypted image in cloud becomes an urgent problem to be solved. Therefore, the study of the encrypted image retrieval has very important significance to the development of cloud computing.Existing encrypted image retrieval technology is seldom used in the cloud, and often do not meet the cloud computing scenarios. This paper proposes the content based encrypted image retrieval technology. The main contributions of this paper are as follows:1) Propose an encrypted image retrieval scheme based on global image feature.This scheme extracts global image features from the image and transforms the global images for security. The transformation of global feature not only protects the privacy of images, but also supports the similarity of images. Extensive experiments also show the feasibility of the proposed scheme. Finally, a security analysis is given in the end of the scheme. 2) Propose an encrypted image retrieval scheme based on local image feature.This scheme extracts local image features from the image. All the local features are put together and clustered. After clustering, every image is represented by the cluster center. And then, the image center is used to construct the searchable index. This scheme detailed describes the process of transforming the distance computation problem to the corresponding linear programming problem. Extensive experiments are carried out to prove the feasibility of the proposed scheme. Finally, the proposed scheme is also proved to be secure with a strict security analysis.