摘要
采用分时变频的采集模式,应用Java媒体框架模型,设计并实现了蛋鸡养殖过程中海量数据自动处理系统。系统既可同步采集和保存关键的环境、图像及音频信息,又能减少存储空间、降低存储成本和加速查询速度。系统运行结果表明:依据采集设备及采集参数设置的不同,单采集点一天采集的音频及图像数据量为200～400 MB,一年采集的数据量为70～140 GB。系统同时提供了以采集时间戳为索引的多源信息同步查询及同步展示模块,可复原大部分生产过程场景。
For providing traceable farm production, a data automatic processing system for egg production was designed and implemented, which altered frequency of acquisition based on time. Java media framework was used as the main model to acquire and process massive image and sound datasets. The system could not only acquire and save synchronized environmental, image and sound data, but also significantly reduce storage space and costs, and speed up query speeds. Tests show that image and sound data in one day is only 200 - 400 MB, and collected data in one year is about 70 - 140 GB, yet contains most of the scenes of the production process.
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
2012年第8期193-197,共5页
Transactions of the Chinese Society for Agricultural Machinery
基金
国家自然科学基金资助项目(31072066)
公益性行业(农业)科研专项经费资助项目(201003011)
关键词
蛋鸡
养殖
图像
音频
海量数据
Laying hens, Raise, Image, Sound, Massive data
