期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2017
卷号：95
期号：23
页码：6635
出版社：Journal of Theoretical and Applied
摘要：Multimedia data security and storage space allocation on cloud servers is a matter of concern for many CSPs and also has a vast scope for research. Media files generally are encrypted and stored on storage servers due to various security threats. Though recent advancements are in mass storage density of servers and high speed processors have provided a little relaxation to CSPs but still with their limited storage capacity and heavy usage of services by users globally, these storage servers usually pave way for high memory allocations for media files resulting in lack of server space. We argue that uncompressed media such as images take more storage space compared to compressed files and also consume more time for cipher operations, which results in poor performance and considerably high bandwidth usage for operations. We propose a combination of Blowfish encryption algorithm with Artificial Neural Networks to provide an efficient way to store and process media files on servers. We have evaluated the performance of proposed work and compared it in terms of PSNR, compression ratio, mean square error, average difference, maximum difference and normalized absolute error and time efficiency during cipher operations. Through these experimental results we prove the efficiency of system which increases dramatically using BANN technique.
