检索规则说明:AND代表“并且”;OR代表“或者”;NOT代表“不包含”;(注意必须大写,运算符两边需空一格)
检 索 范 例 :范例一: (K=图书馆学 OR K=情报学) AND A=范并思 范例二:J=计算机应用与软件 AND (U=C++ OR U=Basic) NOT M=Visual
作 者:谷宇 陈怡帆 邓伟环 李筠筠 李郇[1] Gu Yu;Chen Yifan;Deng Weihuan;Li Junjun;Li Xun(China Regional Coordinated Development and Rural Construction Institute,School of Geography and Planning,Sun Yat-sen University,Guangzhou 510275,Guangdong,China)
机构地区:[1]中山大学地理科学与规划学院中国区域协调发展与乡村建设研究院,广东广州510275
出 处:《地理科学》2025年第1期202-213,共12页Scientia Geographica Sinica
基 金:国家自然科学基金项目(42371206);广东省自然科学基金项目(2023A1515011276)资助。
摘 要:卫星遥感具有覆盖范围大、时空标准统一等优势,为研究传统民居整体性的传承与变化提供了数据基础;然而,当前传统民居研究中的遥感解译主要依赖于人工目视和手工转绘,其深度和效率无法满足需求,如何基于遥感智能解译实现传统民居相似性测度亟待解决。本文提出了一种遥感智能解译的传统民居相似性测度方法,能够融合传统民居专家知识和深度学习模型,以79.1%的总体分类精度和77.1%的表征能力,有效提取传统民居关键特征。在此基础上,以客家民系为视角,测度全国尺度下的客家民居相似性;结合移民史、古驿道和村志记载,分析明清时期南方人口迁徙对客家民居相似性变化的整体性影响,为传统民居和人口迁徙研究提供新视角。The inheritance and development of Chinese vernacular housing and the impact of population migration on these are important topics in the research of vernacular housing.However,due to the spatial and temporal limitations of traditional case surveys,studies conduc-ted from architectural perspectives often lack the ability for inter-field comparisons.As a result,the similarities in the overall spatial patterns of Chinese vernacular housing remain unclear.Satellite remote sensing holds significant potential for addressing these challenges because of its extensive coverage and consistent spatial-temporal criteria.However,in vernacular housing research,remote-sensing interpretations rely heavily on manual processing and drawing,limit-ing the depth and efficiency of interpretation necessary for effectively measuring similarities.Measuring similarities in vernacular housing through intelligent remote-sensing interpretation presents considerable challenges.This paper proposes a method for measuring vernacular hous-ing similarities by utilizing intelligent remote-sensing interpretation in conjunction with large amounts of satellite imagery.This method combines expert knowledge with deep learning mod-els to achieve effective similarity measurement and mapping,resulting in an accuracy of 79.1%for classifying classical vernacular housing and a representative ability of 77.1%across all im-agery features.Using this method,we assessed the vernacular housing similarities from the Hakka perspective throughout China.The effectiveness of feature extraction and similarity measurement was validated by analyzing Hakka migration routes,ancient postal roads,and vil-lage documents.The findings indicate that population migration in southern China during the Ming and Qing dynasties significantly influenced the spatial patterns of Hakka vernacular hous-ing similarities.In conclusion,this study offers a new perspective on vernacular housing and migration research.
关 键 词:中国传统民居 人口迁徙 卫星遥感 深度学习 特征相似性
分 类 号:K901.9[历史地理—人文地理学]
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在链接到云南高校图书馆文献保障联盟下载...
云南高校图书馆联盟文献共享服务平台 版权所有©
您的IP:216.73.216.198