https://github.com/felipeagger/watchouse

Python OpenCV Application to monitor streaming video to detect people and trigger alerts.
https://github.com/felipeagger/watchouse

Last synced: 2 days ago
JSON representation

Python OpenCV Application to monitor streaming video to detect people and trigger alerts.

• Host: GitHub
• URL: https://github.com/felipeagger/watchouse
• Owner: felipeagger
• License: agpl-3.0
• Created: 2023-09-17T19:45:00.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2023-09-17T20:31:53.000Z (about 1 year ago)
• Last Synced: 2023-09-17T21:31:56.355Z (about 1 year ago)
• Language: Python
• Size: 4.18 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# watchouse

Python Application to monitor streaming video to detect people and trigger alerts.

Initially custom to use in a Raspberry Pi and support only RTSP as a Streaming Source.

The script is main.py and the envs is required:

- HOST

- PORT

- USR

- PASSWORD

Personalize the send_alert function to make request to your favorite service trigger a push. (My is Push-bullet)

And in this case the ENV ACCESS_TOKEN is required.

# Dependencies

- Python >= 3.9

- OpenCV (https://opencv.org/)

- YoloV5 Model (https://github.com/ultralytics/yolov5)

- torch

- torchvision

- memoization (to make a memory cache and notify in 5 minute windows)

The model used is the smallest model: models/yolov5n.pt