Scholarlink

<p>The widespread adoption of urban sensing technologies, especially computer vision (CV), has enabled a shift in crowd management from reactive surveillance to proactive monitoring. CV research demonstrates high precision in detecting, counting, and tracking pedestrians in complex urban settings. However, a significant prescriptive gap persists. Despite CV’s growing capability to detect, diagnose, and predict crowd issues, its data remains confined to operational dashboards, failing to systematically inform evidence-based design guidelines for the built environment. This paper addresses this gap by conducting a systematic literature review (<i>n</i> = 27) from an initial sample of 500, followed by a reflexive thematic analysis to formulate a new conceptual framework that advances the traditional workflow into the Detect-Diagnose-Predict-Design (D<sup>2</sup>PD) framework. The new framework provides a structured methodology that enables urban planners to leverage CV analytics as a diagnostic and predictive tool, thereby moving beyond mere monitoring towards generative and evidence-based urban design.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

An iterative framework to inform urban design with computer vision

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月1日

查看全文
  • Alshimaa Aboelmakarem Farag

摘要

The widespread adoption of urban sensing technologies, especially computer vision (CV), has enabled a shift in crowd management from reactive surveillance to proactive monitoring. CV research demonstrates high precision in detecting, counting, and tracking pedestrians in complex urban settings. However, a significant prescriptive gap persists. Despite CV’s growing capability to detect, diagnose, and predict crowd issues, its data remains confined to operational dashboards, failing to systematically inform evidence-based design guidelines for the built environment. This paper addresses this gap by conducting a systematic literature review (n = 27) from an initial sample of 500, followed by a reflexive thematic analysis to formulate a new conceptual framework that advances the traditional workflow into the Detect-Diagnose-Predict-Design (D2PD) framework. The new framework provides a structured methodology that enables urban planners to leverage CV analytics as a diagnostic and predictive tool, thereby moving beyond mere monitoring towards generative and evidence-based urban design.