Improving Infrastructure Monitoring: UAV-Based Photogrammetry for Crack Pattern Inspection

Abstract

Infrastructure degradation, including cracking and occlusions, poses significant risks to structural integrity, demanding efficient
monitoring and interventions. Geomatic techniques, especially UAV photogrammetry, offer promising avenues for crack pattern
inspection. This study aims to develop a rapid and replicable investigation methodology for crack pattern inspection applicable
across various materials and structures. Initially focusing on reinforced concrete, the research aims to optimize the investigation process, favoring automatic or semi-automatic approaches. Exploiting UAV-based photogrammetry, detailed and panoramic images facilitate crack identification and structural health assessment. The methodology involves photogrammetric reconstruction of specimens, orthophoto extraction, and filtering for edge enhancement. Object-Based Image Analysis (OBIA) classification is utilized for automatic crack extraction. The study evaluates the effectiveness of specific filters, including Enhanced Frost and Median, in refining crack detection. While promising results are obtained, further refinement and testing are warranted. The proposed methodology holds the potential for creating a rapid, effective, and easily replicable infrastructure monitoring system, contributing to safety and sustainability.