摘要
污染场地的管控主要依靠统计报告及图像、照片等数据,监测频率低,周期长,效率低,特别是统计报告信息来源单一,影响了管控的精准度和有效性。针对污染场地智能化管控对多源异构数据融合的需求,综述了多源异构数据融合及相关技术,介绍了常见的多源异构数据融合方法,讨论了污染场地多源异构数据融合应用中的难点,展望了未来的研究方向,以期为面向污染场地智能化管控的多源异构数据融合技术研发提供理论基础及方法参考。
Control of contaminated sites mainly relies on statistical reports,images,photos and other data.The monitoring frequency is low,the period is long,and the efficiency is low.In particular,the information source of statistical reports is single,which affects the accuracy and effectiveness of the control.In view of the requirement of multi-source heterogeneous data fusion in the intelligent management and control of contaminated sites,the study summarizes the multi-source heterogeneous data fusion and related technical methods,introduces the common methods of multi-source heterogeneous data fusion,discusses the difficulties in the application of multi-source heterogeneous data fusion in contaminated sites,and prospects the future research orientation,to provide theoretical basis and method reference for the research and development of multi-source heterogeneous data fusion technology based on intelligent management and control of pollutant sites.
作者
陈征
高明亮
蒋卫国
李志涛
Chen Zheng;Gao Mingliang;Jiang Weiguo;Li Zhitao(Technical Centre for Soil,Agriculture and Rural Ecology and Environment,Ministry of Ecology and Environment,Beijing 100012,China;College of Resource Environment and Tourism,Capital Normal University,Beijing 100048,China;Faculty of Geographical Science,Beijing Normal University,Beijing 100875,China;Chinese Research Academy of Environment Sciences,Beijing 100012,China)
出处
《黑龙江科学》
2023年第8期1-6,12,共7页
Heilongjiang Science
基金
国家重点研发计划子课题(2020YFC1807403)。
关键词
特征级融合
深度学习
贝叶斯估计
信息熵
D-S推理
Feature-level data fusion
In-depth learning
Bayesian estimation
Information entropy
D-S reasoning