How sribhavanayalla-01/sitewatch-project works
SiteWatch AI is a Streamlit construction-safety monitoring prototype. Users upload images or videos; the app processes them, detects objects, classifies risks, presents recommendations, and records qualifying incidents for later history and dashboard views. The README documents the processing stages, while the sampled app.py excerpts confirm image analysis, database access, and incident saving; unsampled implementation details are represented only where the README supports them.
Main entry points: app.py YOLO detection · sitewatch.db Incident database · yolo11n.pt YOLO weights
Components 15
User interface
- Streamlit application
app.py - Image upload
app.py - Video upload
app.py
Safety analysis
- Image processing
app.py - Video processing
app.py - YOLO detection
app.py - YOLO weights
yolo11n.pt - Risk assessment
app.py - Safety recommendations
app.py - Detection annotation
app.py
Incident records
- Incident database
sitewatch.db - Incident history
app.py
Safety insights
- Safety dashboard
app.py - Overview page
app.py
Other components
- Site inspector
Connections 15
- Site inspector → Streamlit application: uses
README.md - Streamlit application → Image upload: offers
README.md - Streamlit application → Video upload: offers
README.md - Image upload → Image processing: submits image
README.md - Video upload → Video processing: submits video
README.md - Image processing → YOLO detection: processes with
README.md - Video processing → YOLO detection: analyzes frames
README.md - YOLO detection → YOLO weights: uses weights
README.md - YOLO detection → Risk assessment: provides detections
README.md - Risk assessment → Safety recommendations: informs
README.md - YOLO detection → Detection annotation: draws boxes
README.md - Streamlit application → Incident database: saves incidents
app.py - Incident database → Incident history: supplies records
README.md - Incident database → Safety dashboard: supplies analytics
README.md - Incident database → Overview page: supplies metrics
app.py