Setup

$ workon <env_name> # my environment is named py3cv4
$ pip install opencv-contrib-python
$ pip install zmq
$ pip install imutils

Hardware

• 1x Raspberry pi zero
• 1x Pi camera v2
• 1x Huawei e3372 usb modem
• 1x USB phone battery charger

Setup

1. Configure modem with SIM card
2. Set up Raspberry pi with camera using Motion to run when active and stream endpoint
3. Set up OpenCV machine for analyzing video for faces

Active process

1. Run camera and stream video to server that analyzes stream
2. Identified faces stored as images and embedded
3. Embedded faces added to DB
4. Faces tagged with "session"
5. If photo fits category, trigger alert

---

Links

Reference

• https://github.com/deepinsight/insightface
• https://github.com/tadasbaltrusaitis/openface
• https://github.com/davidsandberg/facenet
• http://shuoyang1213.me/WIDERFACE/WiderFace_Results.html

Inspiration

• https://www.nytimes.com/2019/04/16/opinion/privacy-technology.html
• https://www.nytimes.com/2019/04/17/opinion/data-privacy.html
• https://www.nytimes.com/interactive/2019/04/16/opinion/facial-recognition-new-york-city.html
• https://cvdazzle.com/

Code

• https://www.pyimagesearch.com/2019/04/15/live-video-streaming-over-network-with-opencv-and-imagezmq/