摘要
从江苏省养蚕、制种生产环境调控和物联网技术应用的现状出发,针对现有农业模式的弊端以及农业物联网控制环节的滞后性,设计一种基于深度学习的物联网蚕室环境管理系统。利用神经网络结合CNN深度学习算法对数据样本不断学习训练,除了实现远程智能控制、信息共享外,最重要的是不仅实现了对蚕桑环境因子的进一步精确整定,还提高了整个系统的快速响应能力以及鲁棒性。
Starting from the current situation of sericulture,seed production environment regulation and the application of IoT in Jiangsu Province,aiming at the drawbacks of the existing agricultural model and the lag of the control link of the IoT in agriculture,this paper designed an IoT silkworm room environment management system based on deep learning.We used neural network combined with CNN deep learning algorithm to train data samples continuously to realize remote intelligent control and information sharing.The most important thing is not only to achieve further precise tuning of environmental factors in silkworm room,but also to improve the rapid response ability and robustness of the whole system.
作者
孙孝龙
吴洪昊
周卫阳
施建军
Sun Xiaolong;Wu Honghao;Zhou Weiyang;Shi Jianjun(Yancheng Bioengineering Branch of Jiangsu Union Technical Institute, Yancheng 224051, Jiangsu, China;Jiangsu Silkworm Breeding Management Office, Wuxi 214151, Jiangsu, China;Yancheng Siyuan Network Technology Co., Ltd., Yancheng 224005, Jiangsu, China)
出处
《计算机应用与软件》
北大核心
2019年第6期58-61,75,共5页
Computer Applications and Software
基金
国家职业教育数字化资源共建共享计划项目(ZYWZ201101)
科技部“十二五”农村领域科技计划项目(2013GA690067)
江苏省“333”人才培养工程2014年项目(BRA2014318)
江苏省科技厅2010年农业科技支撑计划项目(BE2010304)
江苏省2015年农业三新工程项目(SXGC[2015]187)
关键词
物联网
蚕室环境
管理系统
深度学习
IoT
Silkworm environment
Management system
Deep leaning
