基于零参考深度曲线估计的图像增强网络改进

随着多云存储市场的快速发展, 越来越多的用户选择将数据存储在云上, 随之而来的是云环境中的重复数据也呈爆炸式增长. 由于云服务代理是相互独立的, 因此传统的数据去重只能消除代理本身管理的几个云服务器上的冗余数据. 为了进一步提高云环境中数据去重的力度, 本文提出了一种多代理联合去重方案. 通过区块链技术促成云服务代理间的合作, 并构建代理联盟, 将数据去重的范围从单个代理管理的云扩大到多代理管理的多云. 同时, 能够为用户、云服务代理和云服务提供商带来利益上的共赢. 实验表明, 多代理联合去重方案可以显著提高数据去重效果、节约网络带宽.

Improved Image Enhancement Network Based on Zero Reference Deep Curve Estimation

With the rapid development of the multi-cloud storage market, more and more users choose to store data in the cloud, followed by the explosive growth of duplicate data in the cloud environment. Because cloud service brokerages are independent of each other, traditional data deduplication can only eliminate redundant data on several cloud servers managed by the brokerages themselves. To further improve the data deduplication in a cloud environment, this paper proposes a multi-brokerage joint deduplication scheme. Through the blockchain technology to promote the cooperation between cloud service brokerages and the construction of a brokerage alliance, the scope of data deduplication is extended from single brokerage managed clouds to multiple brokerages managed clouds. At the same time, it can bring win-win benefits to users, cloud service brokerages and cloud service providers. Experiments show that the multi-brokerage joint deduplication scheme can significantly improve the data deduplication effect and save network bandwidth.