Skip to content
SDGs Center UDINUS

Evidence record · UDINUS · SDGs

Asphalt Road Quality Classification Based on CNN Architecture VGG16

Rektorat UDINUS 2023 Published 2026-10-30

Evidence record · UDINUS · SDGs

Unit
Rektorat UDINUS
Period
2023
Goals
Record
a5fb6c9f-a2ab-4193-8a1b-7ff852c8de43
Published
2026-10-30

Universitas Dian Nuswantoro conducted research on the classification of asphalt highway quality using a convolutional neural network method with the VGG16 architecture. The activity involved Utomo, Sapto; Wachid Affandi, Mochamad Noor; and Fajar Shidik, Guruh as research participants. The study addressed damaged road conditions, which often cause accidents and may result in casualties and property loss when vehicles become stuck, are struck, or collide with another vehicle while avoiding damaged road sections.

The research examined the classification of good and bad asphalt highway quality. This classification was presented in the context of surveying asphalt road conditions for the transportation agency. The source explains that such surveying is needed to ensure road safety and to support the proper implementation of road maintenance and rehabilitation. The activity therefore used image-based classification to distinguish between good and bad asphalt highway conditions through the VGG16 architecture.

Two tests were conducted. The first test used images measuring 224x224, while the second test used images measuring 64x64. The test using 224x224 images obtained an accuracy value of 92.47%. The test using 64x64 images obtained an accuracy level that exceeded the 87.80% result reported in research [12]. These findings describe the reported performance of the VGG16-based convolutional neural network in classifying asphalt highway quality under the two tested image sizes. UDINUS frames the activity as research addressing asphalt road-condition classification, road safety, and the implementation of road maintenance and rehabilitation.

Reference https://www.scopus.com/pages/publications/85177996504?origin=resultslist