Distance-Driven Pavement Rating with UAV Imagery
Authors/Creators
Description
Pavement rating systems such as the Pavement Condition Index (PCI) and Pavement Surface Evaluation and Rating (PASER) combine severity and spatial distribution into composite scores. This paper introduces the Distance-Based Pavement Condition Index (DBPCI), a rule-based Unmanned Aerial Vehicle (UAV)-compatible metric that derives a 1–10 rating from inter-defect spacing in UAV imagery. Unlike traditional indices, DBPCI excludes severity weighting, isolating spacing as an independent indicator of condition. DBPCI scoring is rule-based, while machine learning is used only for defect detection. Validation compared machine learning predicted ratings with expert assigned PASER ratings across merged maintenance categories. DBPCI achieved 100% accuracy in three categories (1–2, 5–7, 8–10) and 70% accuracy in one category (3–4), demonstrating that defect spacing alone provides a reliable basis for pavement condition assessment. DBPCI provides a rapid, distance-focused complementary index to existing indices, enhancing UAV-based pavement management by identifying segments requiring priority maintenance.