IROSCD: Indonesia Road Surface Classification Dataset
Description
Road surface condition is a critical factor in transportation safety, vehicle maintenance, and the development of intelligent transportation systems. Damaged, uneven, or wet road surfaces can increase the risk of accidents and reduce driving comfort. However, research on road surface classification in Indonesia remains limited due to the absence of a representative and realistic dataset reflecting local road conditions. To address this gap, we introduce the Indonesia Road Surface Classification Dataset (IROSCD), a new dataset that captures real road conditions in Indonesia under four categories: Normal Road, Damaged Road, Bumpy Road, and Wet Road. Each category includes typical visual characteristics such as cracks, potholes, bumps, and water puddles, collected from various regions across Indonesia. The IROSCD dataset is expected to serve as a benchmark for further studies in intelligent transportation systems , road condition monitoring , and autonomous vehicle research , both within Indonesia and globally.
Files
Steps to reproduce
Images were captured using smartphone in various urban and rural locations across Indonesia, covering different environmental and road conditions. Each image was manually reviewed to ensure label consistency and visual quality.
Institutions
- Badan Riset dan Inovasi Nasional Republik IndonesiaJakarta, Central Jakarta