基于机器学习和街景图像的历史城区店招色彩量化研究——以苏州为例  

Quantifying the Color of Signboards in Historic District Based on Machine Learning and Street View Images:A Case Study of Suzhou

作　　者：朱隆斌[1] 赵瑞寅 ZHU Longbin;ZHAO Ruiyin

机构地区：[1]南京工业大学建筑学院

出　　处：《现代城市研究》2024年第6期120-125,共6页Modern Urban Research

摘　　要：店招色彩是城市风貌的重要组成元素,特别是风貌重点管控的历史城区。文章运用机器学习技术和街景图像数据,以苏州历史城区为例,基于“传统建筑一管控建筑一现代建筑”3个尺度构建一套历史城区店招色彩的评价方法。该方法以街道为单位识别不同类型建筑的店招主导色彩并对街道的店招色彩协调度进行评价。通过筛选协调度较高和较低的街道,分析其店招色彩特点,提出历史城区店招色彩量化管控的依据。该方法可为历史城区店招的精细化更新管控提供技术支持。The color of signboards is an important component of urban landscape,especially in the historic district where the landscape is primarily controlled.Using machine learning technology and street view image data,this paper constructs a method for evaluating the color of signboards in the historic district of Suzhou based on three categories:"traditional buildings-controlled buildings-modern buildings",The method identifies the dominant colors of signboards of different types of buildings by street and evaluates the color coordination of signboards on each street.By screening the streets with higher and lower coordination,the color characteristics of their signboards are analyzed and the basis for quantitative color control of the signboards in the historic district is proposed.The method can provide technical support for the refined renewal of signboards in historic districts.

关 键 词：街景数据 店招色彩 色彩评价 历史城区 机器学习 

分 类 号：TU984[建筑科学—城市规划与设计]