A Correction Algorithm for Automatic Pavement Detection Data Bias

Results from automated pavement inspection equipment often deviate from the pavement’s actual condition, which leads to misjudgment of the pavement’s technical condition. This study takes inspection data obtained by automated equipment commonly used on Shanghai highways as an example and proposes a data correction process that considers the pavement damage composition characteristics. Based on 10.8 km of automated equipment and manual comparative inspection testing, the study suggests a data correction algorithm based on the single damage index and establishes a PCI (Pavement Condition Index) calculation model based on automated inspection data that integrates block crack, alligator crack, line crack, and the interaction term. Finally, the proposed correction model is validated with actual pavement measurement data, and the results show that the model has good accuracy.