城市建筑2025，Vol.22Issue(22)：34-36,3.DOI:10.19892/j.cnki.csjz.2025.22.10
基于多源大数据及深度学习的广州市中心城区街道品质测度
The Measurement of Street Quality in the Urban Central Area of Guangzhou Based on Multi-Source Big Data and Deep Learning Algorithms
摘要
Abstract
This paper employs crowdsourced geographic infor-mation big data and deep learning algorithms to measure street quality and provides a detailed spatial characterization of streets.The research shows that:① the efficiency evaluation of the streets in the main urban area of Guangzhou exhibits a single efficient characteristic,either in terms of transportation convenience or func-tional completeness.② Guangzhou's streets have a high coverage of pedestrian transportation facilities,but safety measurements show significant disparities.③ The core areas of the urban central areas have high enclosure levels,resulting in distinct spatial heterogeneity in street comfort evaluations.This paper highlights the importance of street functional diversity and the diversification of travelers'needs,providing new methods and new ideas for achieving refined urban governance.关键词
街道品质/多源数据/街景图片/深度学习Key words
street quality/multi-source data/street view images/deep learning分类
建筑与水利引用本文复制引用
李莹,章礼杰,卢智婷..基于多源大数据及深度学习的广州市中心城区街道品质测度[J].城市建筑,2025,22(22):34-36,3.基金项目
广州市科技局广州市资源规划和海洋科技协同创新中心项目"空天地海一体化自然资源智能监测评价技术研究"(2023B04J0301,2023B04J0326) （2023B04J0301,2023B04J0326）
