How divaprof0912-cyber/traffic-violation-detection-system works
The product is a Streamlit image-based traffic-violation detector. A user uploads an image; the app converts it to an array, runs YOLO inference, checks detections for helmet violations or triple riding, and displays an annotated image or a no-violation warning. The README also describes vehicle–violation association using IoU, but the sampled app excerpt does not show that step. The runtime map is based on app.py; no other source modules appear in the tree.
Main entry points: app.py YOLO Inference · best.pt Model Weights
Components 7
User Interface
- Image Upload
app.py - Result Display
app.py
Detection Workflow
- Image Conversion
app.py - YOLO Inference
app.py - Model Weights
best.pt - Violation Check
app.py
Other components
- User
Connections 6
- User → Image Upload: uploads image
app.py - Image Upload → Image Conversion: provides image
app.py - Image Conversion → YOLO Inference: passes array
app.py - YOLO Inference → Model Weights: loads weights
app.py - YOLO Inference → Violation Check: returns detections
app.py - Violation Check → Result Display: shows outcome
app.py