metadata
license: mit
language:
- en
pipeline_tag: image-classification
tags:
- computer-vision
- accident-detection
- yolov11
Accident Classifier
Model Description
This model classifies road-scene images into two categories: accident and no_accident. It is part of an academic AI-Based Road Accident Detection and Emergency Report Generation System.
Intended Use
- Road-scene image classification for academic demonstrations.
- Supporting a prototype accident-analysis workflow.
- Educational experimentation with computer vision and deep learning.
Model Details
- Architecture: YOLO11n classification model
- Task: Binary image classification
- Classes:
accident,no_accident - Framework: Python, Ultralytics, PyTorch
Evaluation
- Test accuracy: 98.12%
- Test images: 1,858
Results are based on the project's current held-out test dataset. Performance on new, real-world images may differ.
Limitations and Safety
This model is an academic prototype, not a verified emergency-response system. It may misclassify images and must not be used as the sole basis for emergency decisions. Predictions should be reviewed by a human. It does not automatically contact emergency services.
Related Project
The model is used in the AI-Based Road Accident Detection & Emergency Report Generation System.