摘要
基于京雄城际铁路施工期生态环境影响的分析,通过对大型临时工程和环境敏感点遥感影像的切片、样本制作等预处理,研究建立了基于深度卷积神经网络的分类识别模型,使用优化算法对模型进行模拟训练,基本可实现对铁路沿线特定目标的智能识别。该技术的应用为京雄城际铁路开展生态监控提供了有力支撑,促进信息化、自动化、智能化铁路建设。
Based on the analysis of the eco-environmental impact of the construction period of Beijing-Xiong,an Intercity Railway,a classification and recognition model based on deep convolution neural network is established through the pretreatment of remote sensing image slice and sample production of large-scale temporary works and environmental sensitive points.The model is simulated and trained with optimization algorithm,and the intelligent recognition of specific targets along the railway is basically realized.The application of this technology provides a strong support for the ecological monitoring of Beijing-Xiong,an Intercity Railway and promotes the construction of informatization,automation and intelligence railway.
作者
葛辉凯
郝爽
葛开国
GE Huikai;HAO Shuang;GE Kaiguo(Bridge Department,Wuhan Railway SiYuan Engineering Consulting Co.,Ltd.,Wuhan Hubei 430063,China;Xiong,an Headquarters,Xiong,an High Speed Railway Co.,Ltd.,Shijiazhuang Hebei 050000,China;Energy Saving&Environmental Protection&Occupation Safety and Health Research Institute,China Academy of Railway Sciences Co.,Ltd.,Beijing 100081,China)
出处
《铁路节能环保与安全卫生》
2022年第1期1-5,共5页
Railway Energy Saving & Environmental Protection & Occupational Safety and Health
关键词
京雄城际铁路
生态环境
卷积神经网络
分类识别
智能化
Beijing-Xiong,an Intercity Railway
Ecological environment
Convolution neural network
Classification and recognition
Intelligentize
