Scholarlink

<p>Infrastructure inspection is essential for ensuring the safety and operational performance of public works. However, many public agencies in northern Brazil face limitations related to procedures, equipment, and qualified personnel, which restrict effective supervision. This article proposes and validates a standardized, low-cost methodology for infrastructure inspection using commercially available drones, scale bars, photogrammetric processing, and artificial intelligence-based diagnostic support. The methodological protocol integrated structured flight planning, image acquisition under controlled photometric conditions, scale bar calibration, three-dimensional reconstruction in Agisoft Metashape, orthomosaic generation, and dimensional validation against manual field measurements. Four case studies conducted in Gold Coast, Australia, including urban pavement, a reinforced concrete bridge, a commercial building façade, and a reinforced concrete dam, were used to assess reproducibility and precision. The results demonstrated sub-millimetric calibration error (&lt; 1&#xa0;mm), sub-centimeter measurement accuracy, and consistent detection of structural pathologies when compared with benchmark field measurements. The use of scale bars eliminated the need for ground control points and reduced equipment costs to approximately $500, representing up to 80% savings compared to RTK-based systems. The findings indicate that the proposed methodology provides technical reliability, economic feasibility, and operational accessibility, enabling public agencies, especially in remote and resource-constrained regions, to perform systematic infrastructure inspections. This framework contributes to the democratization of advanced inspection technologies and offers a scalable model for construction pathology assessment and public infrastructure management.</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

Public works inspection: a low-cost, AI-enhanced drone methodology for remote communities

  • 期刊论文

  • 发表日期：

    2026年2月25日

查看全文
  • Thiago Dias de Araújo e Silvaa,
  • Fernanda Helfer

摘要

Infrastructure inspection is essential for ensuring the safety and operational performance of public works. However, many public agencies in northern Brazil face limitations related to procedures, equipment, and qualified personnel, which restrict effective supervision. This article proposes and validates a standardized, low-cost methodology for infrastructure inspection using commercially available drones, scale bars, photogrammetric processing, and artificial intelligence-based diagnostic support. The methodological protocol integrated structured flight planning, image acquisition under controlled photometric conditions, scale bar calibration, three-dimensional reconstruction in Agisoft Metashape, orthomosaic generation, and dimensional validation against manual field measurements. Four case studies conducted in Gold Coast, Australia, including urban pavement, a reinforced concrete bridge, a commercial building façade, and a reinforced concrete dam, were used to assess reproducibility and precision. The results demonstrated sub-millimetric calibration error (< 1 mm), sub-centimeter measurement accuracy, and consistent detection of structural pathologies when compared with benchmark field measurements. The use of scale bars eliminated the need for ground control points and reduced equipment costs to approximately $500, representing up to 80% savings compared to RTK-based systems. The findings indicate that the proposed methodology provides technical reliability, economic feasibility, and operational accessibility, enabling public agencies, especially in remote and resource-constrained regions, to perform systematic infrastructure inspections. This framework contributes to the democratization of advanced inspection technologies and offers a scalable model for construction pathology assessment and public infrastructure management.