Inspection system for the conservation of flexible pavements through the Cascade Trainer algorithm and the use of the DJI Mini 4 Pro drone

The pavement inspection process is of utmost importance to understand its condition and thus make decisions aimed at its preservation and maintenance. This research proposes a methodology to optimize the pavement inspection process based on the ASTM D6433 standard through the use of drones and the Cascade Trainer algorithm. Images of pavement pathologies were used to train the algorithm and perform pavement inspections in videos captured by the DJI mini 4 pro drone. The algorithm demonstrated high performance with an average accuracy of 96%, a 77% cost reduction compared to the "American Society for Testing and Materials" (ASTM D6433) method, and a performance capable of measuring 600 additional meters per working day. Finally, the results of this research may be useful to promote a proposal to improve flexible pavement inspection for preservation through automated and highly accurate methods.