intelligent_surveillence_system

A terible multi-threaded pyqt5 application that uses custom trained yolov4-tiny model to detect 4 different classes of weapons and sends email and sms about first detection, email includes images.

UI

Restart button hasnot been programmed

To run locally

1. Create a virtual enviromnet and run pip install -r requirements.txt
2. Add .env file on root of project and add following on files

# email
EMAIL_AUTH_TOKEN="your_token"
# SMS
SMS_ACCOUNT_SID="your_twilio_sid"
SMS_AUTH_TOKEN="your_twilio_auth_token"

3. Update credentials.json to add following

"EMAIL_SENDER": "your_email_sender_account",
"EMAIL_RECEIVERS": [
  "email_1",
  "email_2",
]

"SMS_SENDER":  "your_sender_twilio_number",
"SMS_TARGETS": ["receiver_1",]

4. Make directory to stored detected images mkdir images
5. Run python app.py