How navjyotsingh17/aadhaar_masking works
This repository is an Aadhaar-document masking service: a client submits document bytes to the Flask API, which checks the supplied password and delegates masking to the image-processing engine before returning status and, when masked, document bytes. The API also exposes classification and audit-related routes. Audit queries and logging configuration are visible, but the excerpts do not establish every route’s behavior or the engine’s detailed OCR/detection call sequence; those internals are therefore kept consolidated.
Main entry points: app.py Flask API · wsgi.py WSGI server · mask_image.py Masking engine · models Model weights
Components 8
Service runtime
- WSGI server
wsgi.py - Flask API
app/app.py
Document processing
- Masking engine
services/mask_image.py - Model weights
models
Audit and logs
- Audit database
app/app.py - Log setup
config/logging_config.py
Request security
- Password crypto
utils/encrypt_decrypt.py
Other components
- Document client
Connections 8
- Document client → WSGI server: submits request
Readme.md - WSGI server → Flask API: serves app
wsgi.py - Flask API → Password crypto: decrypts password
app/app.py - Flask API → Masking engine: delegates masking
app/app.py - Masking engine → Model weights: loads weights
services/mask_image.py - Flask API → Audit database: queries audit data
app/app.py - Flask API → Log setup: configures logging
app/app.py - Flask API → Document client: returns result