Deteksi Kerusakan Jalan Berdasarkan Citra Digital Menggunakan Convolutional Neural Network (CNN)

Mulyana, Dadang Iskandar
Wahyudi, Ilham

Jurnal Indonesia : Manajemen Informatika dan Komunikasi
Vol. 6 No. 1 (2025): Januari

Highways are a connection between an area or region to another destination. The rapid construction of highways in big cities is not comparable to the improvement and rearrangement of damaged roads in several areas. Most of the damaged roads are caused by heavy vehicle traffic or heavy loads with quite frequent intensity, as well as natural disasters such as floods and earthquakes. This of course disrupts the traffic system, and is quite dangerous for drivers who often pass through areas where there are many damaged roads. With these obstacles, this study aims to build a system that can detect road damage through digital image capture using the convolutional neural network method. The results of this study obtained a road damage detection accuracy value reaching 80%.