欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
FIFO、LRU、LFU、GDLVF等传统瓦片缓存置换算法侧重于瓦片访问时间和频率、瓦片大小、空间位置关系,不适合具有多源、异构特点的五层十五级瓦片数据,在五层十五级瓦片数据缓存的应用上存在局限性.提出了用户行为参与的缓存置换算法UPBA（User Preference Based Tile Cache Replacement Algorithm）,并从用户行为、瓦片访问的时间和频率、瓦片大小、空间位置关系等方面分析了UPBA算法,提出了提高置换效率的方法.并对最高分辨率为100和250m、生产时间为2014年11月、2015年1~3月的高分一号、高分二号、资源三号影像数据集进行日志驱动仿真实验.结果表明：相较传统的缓存置换算法,UPBA提高了瓦片请求的命中率和字节命中率,降低了客户端流量消耗和服务器端负载.
英文摘要：
Traditional tile cache replacement algorithms such as FIFO,LRU,LFU,GDLVF focusing on tile＇s access time,access frequency,size,and spatial location relationship,have limitations in practice of caching the Five-layer Fifteen-level tile,and are not suitable for the Five-layer Fifteen-level tile which is multi-source heterogeneous.In this paper,a tile cache replacement algorithm for Five-layer Fifteen-level named UPBA（User Preference Based Tile Cache Replacement Algorithm）was proposed,and its features including user preference,tile access time and frequency,tile size,and spatial location relationship were analyzed.And then,the enhanced tile replacement method of UPBA was presented.The image datasets of GF-1,GF-2and ZY-3with resolution of 100 and 250m,and production time in November of 2014 and January,February,March of 2015 were used in log-driven simulations of UPBA.The result showed that the UPBA had improved the request hit rate and byte hit rate,meanwhile reduced the client traffic consumption and the server load compared to the traditional cache replacement algorithm.
同期刊论文项目
同项目期刊论文
Segmentation of High Spatial Resolution Remote Sensing Imagery Based on Hard-Boundary Constraint and
Refining High Spatial Resolution Remote Sensing Image Segmentation for Man-made Objects through aCol
期刊信息
位置：