摘要
新时期城市设计的人本化转型,催生了对于城市尺度下建筑色彩的一系列导控实践。研究针对以往建筑色彩分析难以在大尺度下高效开展的问题,结合街景数据和机器学习,提出了人本视角下的大规模、精细化分析方法。解决了以往建筑色彩分析拘泥于手工小样本采集,难以大范围运用的问题,可快速实现建筑色彩分析与主导色提取,为相关实践提供科学化建议。该分析方法运用开源易得的街景数据,实现了建筑色彩这一难以测度要素的精细化度量,在实践中具有较好的普适性和易用性。
The humanistic transformation of the urban design in the new era has given birth to a series of guiding and controlling practices for architectural color at city scale.In order to solve the problem that the previous architectural color analysis is difficult to carry out at a large scale,combining with street view data and machine learning,this article proposes a large-scale and refined analysis method from humanistic perspective.It solves the problem that the previous architectural color analysis is limited to manual small sample collection,difficult to be used in a wide range,can quickly realize architectural color analysis and dominant color extraction,and provide scientific suggestions for relevant practices.This analysis method uses open source and easy-to-obtain street view data to realize the refined measurement of architectural color,which is difficult to be measured,and has good universality and ease of use in practice.
作者
叶宇
仲腾
钟秀明
YE Yu;ZHONG Teng;ZHONG Xiuming
出处
《住宅科技》
2019年第5期7-12,共6页
Housing Science
基金
国家自然科学基金青年项目(编号:51708410)
国家自然科学基金重点项目(编号:51838002)
上海市浦江人才计划(编号:17PGC107)
关键词
建筑色彩
街景数据
机器学习
人本视角
architectural color
street view data
machine learning
humanistic perspective
