https://github.com/ronibandini/tiam62aitrafficlight
AI driven traffic lights with Texas Instruments AM62A and Edge Impulse
https://github.com/ronibandini/tiam62aitrafficlight
am62a am62x edgeimpulse honda php python texas-instruments unihiker
Last synced: 24 days ago
JSON representation
AI driven traffic lights with Texas Instruments AM62A and Edge Impulse
- Host: GitHub
- URL: https://github.com/ronibandini/tiam62aitrafficlight
- Owner: ronibandini
- License: mit
- Created: 2023-10-20T13:38:55.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2023-10-30T20:46:38.000Z (over 2 years ago)
- Last Synced: 2025-01-20T23:13:07.176Z (over 1 year ago)
- Topics: am62a, am62x, edgeimpulse, honda, php, python, texas-instruments, unihiker
- Language: Python
- Homepage: https://bandini.medium.com/sem%C3%A1foro-con-detecci%C3%B3n-de-cascos-v%C3%ADa-machine-learning-15a18f60d218
- Size: 97.7 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# TIAM62AITrafficLight
AI traffic lights with Texas Instruments AM62A and Edge Impulse.
A model is trained to recognize motorcycle riders with and without helmets. That model is loaded into TI AM62A. A regular webcam will send pictures for local inference. Inference values will be sent to a server.
A separate WiFi device - Unihiker with Debian/Python - will query the server to determine if red traffic light will switch to green.
# Parts
Texas Instruments AM62A
Unihiker board
USB Webcam
Web server/PHP
# Tutorial
https://docs.edgeimpulse.com/experts/image-projects/motorcycle-helmet-detection-smart-light-ti-am62a
# Machine Learning model
https://studio.edgeimpulse.com/public/295740/latest (Trained with Lego)
# 3d printed part
https://www.thingiverse.com/thing:6277702
# Spanish info
https://bandini.medium.com/sem%C3%A1foro-con-detecci%C3%B3n-de-cascos-v%C3%ADa-machine-learning-15a18f60d218
# Demo
https://www.youtube.com/watch?v=YR2Mjf_unmU
# Contact
@RoniBandini