摘要
针对温室种植自动化程度低的情况,提出利用传感器采集多源种植环境数据,结合机器学习方法,进行基于集成学习的温室育种智能决策,不仅可以融合多源种植信息,还可以自适应调整各信息源的权重。实验表明,所提方法可提高温室育种成活率。
Extensive management of China's traditional agriculture leads to low growing efficiency of crops,especially green-house breeding.It will give rise to plant production and influence the development of agriculture.According to low automation of green-house breeding,an intelligence decision algorithm of greenhouse breeding based on ensemble learning is proposed in this paper.Firstly,multi-sources planting environment data are collected by sensors.Secondly,inspired by machine learning,intelligence decision algorithm is designed.It cannot only fuse multi-sources planting information,but also adjust the weights of multi-sources adaptively.Finally,the problem of intelligence decision can be solved.The experiment results indicate that proposed algorithm can provide intelligence decision of green-house breeding,and improve survival rate,which is of great realistic significance for the efficiency of the agricultural economy.
出处
《西南科技大学学报》
CAS
2017年第4期78-81,共4页
Journal of Southwest University of Science and Technology
基金
陕西省协同创新计划项目(2015XT-21)
关键词
智能决策
多源信息
集成学习
温室育种
Intelligence decision
Multi-source information
Ensemble learning
Green-house breeding