地理要素分类机器学习方法发展与前景  

Development and prospects of machine learning methods in geographic elements classification

作　　者：王卷乐[1,3,5] 李凯 严欣荣 郑莉[1,4] 韩雪华 WANG Juanle;LI Kai;YAN Xinrong;ZHENG Li;HAN Xuehua(State Key Laboratory of Resources and Environmental Information System,Institute of Geographic Sciences and Natural Resources Research,Chinese Academy of Sciences,Beijing 100101,China;College of Geoscience and Surveying Engineering,China University of Mining&Technology(Beijing),Beijing 100083,China;University of Chinese Academy of Sciences,Beijing 100049,China;Institute of Disaster Prevention,Sanhe 065201,China;Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application,Nanjing 210023,China)

机构地区：[1]中国科学院地理科学与资源研究所资源与环境信息系统国家重点实验室,北京100101 [2]中国矿业大学(北京)地球科学与测绘工程学院,北京100083 [3]中国科学院大学,北京100049 [4]防灾科技学院,三河065201 [5]江苏省地理信息资源开发与利用协同创新中心,南京210023

出　　处：《遥感学报》2023年第8期1757-1768,共12页NATIONAL REMOTE SENSING BULLETIN

基　　金：中国科学院战略性先导专项(A类)(编号:XDA19040501);中国工程科技知识中心建设项目(编号:CKCEST-2022-1-41)。

摘　　要：地理要素一般包括自然和人文两类对象。日益增加的遥感大数据和泛在的社交媒体数据为这两类对象的要素分类提供了丰富的数据源。基于遥感影像分类为主的自然要素提取和基于网络文本和社交媒体的人文要素提取,是当前地理要素分类的两大主流。前者以图像处理技术为支撑,后者则以自然语言处理技术为核心。随着机器学习等人工智能分类方法的介入,两类要素分类呈现越来越多的共性相通特点。本文以机器学习方法的演变历程为媒介,剖析了其在自然地理要素遥感影像分类和人文社会要素网络文本分类方面的异同。以遥感单一对象、复合对象分类和微博社交媒体话题分类提取为实例,指出二者在机器学习分类方法上具有相通性。遥感大数据和网络文本大数据分类方法的相互借鉴能够促进自然与人文地理要素的智能分类应用。Geographic objects typically include both physical and human elements.The big data produced by remote sensing and the ubiquitous social media data provide rich sources for the feature classification of these two types of objects.The extraction of physical objects based on remote sensing classification and the extraction and classification of social media information based on web text are the current mainstream methods of extracting geographic objects.The former is supported by image processing technology,whereas the latter is achieved using natural language processing technology.With the application of artificial intelligence classification methods such as machine learning,the classification characteristics of these two types of elements are becoming increasingly common.Using the evolution of machine learning methods as a medium,in this study,we compared the remote sensing classification of single-and multiple-element physical geographic elements and the natural language processing classification of web text elements.Since the 1940s,the development of machine learning methods has experienced five stages,namely,germination,development,bottleneck,recovery,and outbreak.Machine learning and related information classification methods have become the current focus of researchers.We described the principle applied by machine learning methods for geographic element classification and divided the classification process of geographic elements into data acquisition,data preprocessing,feature construction or model training,and accuracy evaluation.We observed many similarities between physical-element-oriented remote sensing classification and human-element-oriented text classification in terms of their process and model.However,text and remote sensing classifications also differ in their data and tasks.By using single objects,compound object classification,and microblog social media topic classification extraction as three examples for remote sensing classification,we further examined the process of completing different geog

关 键 词：地理要素分类 自然地理要素 人文地理要素 机器学习 遥感分类 网络文本分类 自然语言处理 

分 类 号：P2[天文地球—测绘科学与技术]