EUVIMEDEUVIMED entdecken
Crossref

Traffic Accident Image Classification using CLIP with a Zero-Shot Approach

Della Septiani, Hartiti Fadilah, Josua Alfa Viando, Anissa Luthfi Alifia, Syifa Firnanda, Pramudya Wibowo, Ade Lailani

Indonesian Journal of Data Risk Research · 15.06.2026 · 1 (1) · 12-25

DOI: 10.55749/ijdrr.v1i1.185 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Indonesian Journal of Data Risk Research, online erschienen am 15.06.2026. Verfasst von Della Septiani und 6 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accident image detection is a critical step in supporting rapid response systems during emergency situations. This study employs the Contrastive Language–Image Pre-training (CLIP) model to detect accident images using a zero-shot approach, without requiring retraining on a specific dataset. The CLIP model leverages multimodal embeddings from text and images, enabling detection based on textual descriptions. The experimental results, using the ViT base patch 32 model, show that this method achieves a Top-1 accuracy of 32% and a Top-5 accuracy of 95.86%. Although the Top-1 accuracy indicates that further optimization is needed, the high Top-5 accuracy demonstrates the significant potential of CLIP for efficient accident image detection. With further development, this technology can serve as a reliable solution for various emergency response scenarios, offering flexibility and efficiency in detecting accident-related images.

Zitieren

APA

Septiani, D., Fadilah, H., Viando, J. A., Alifia, A. L., Firnanda, S., Wibowo, P., & Lailani, A. (2026). Traffic Accident Image Classification using CLIP with a Zero-Shot Approach. Indonesian Journal of Data Risk Research, 1(1), 12-25. https://doi.org/10.55749/ijdrr.v1i1.185

Vancouver

Septiani D, Fadilah H, Viando JA, Alifia AL, Firnanda S, Wibowo P, et al. Traffic Accident Image Classification using CLIP with a Zero-Shot Approach. Indonesian Journal of Data Risk Research. 2026;1(1):12-25. doi:10.55749/ijdrr.v1i1.185

BibTeX
@article{Septiani2026,
  author = {Septiani, Della and Fadilah, Hartiti and Viando, Josua Alfa and Alifia, Anissa Luthfi and Firnanda, Syifa and Wibowo, Pramudya and Lailani, Ade},
  title = {Traffic Accident Image Classification using CLIP with a Zero-Shot Approach},
  journal = {Indonesian Journal of Data Risk Research},
  year = {2026},
  volume = {1},
  number = {1},
  pages = {12--25},
  doi = {10.55749/ijdrr.v1i1.185}
}
Kostenlos starten →