https://github.com/mohammad-nour-alawad/automatic-license-plate-recognition
Python app for ALPR System for Syrian plates, run on NPU
https://github.com/mohammad-nour-alawad/automatic-license-plate-recognition
cnn deep-learning edge-computing iot ip-camera khadas-vim3 object-detection real-time yolo
Last synced: about 1 month ago
JSON representation
Python app for ALPR System for Syrian plates, run on NPU
- Host: GitHub
- URL: https://github.com/mohammad-nour-alawad/automatic-license-plate-recognition
- Owner: mohammad-nour-alawad
- Created: 2023-04-28T11:45:51.000Z (about 2 years ago)
- Default Branch: master
- Last Pushed: 2024-07-10T08:06:12.000Z (10 months ago)
- Last Synced: 2025-02-01T17:12:57.424Z (3 months ago)
- Topics: cnn, deep-learning, edge-computing, iot, ip-camera, khadas-vim3, object-detection, real-time, yolo
- Language: Python
- Homepage:
- Size: 13.7 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Automatic License Plate Recognition (ALPR) for Indian Numbers (٠١٢٣٤٥٦٧٨٩)
THIS PROJECT IS MADE TO WORK ON A **KHADAS VIM3 AI EDGE DEVICE**.
Achived a REALTIME processing at 21 frames per second.
In this project I am deploying a YOLOV4 model for License Plate Recognition of the Indian Numbers (The ones used in the Arabic Language).
**NOTE**: I also send a stream of the detection results UDP port, so I can catch these packets and decode it to see a LIVE STREAM of the detection results.
The Process:
- find the car.
- find the numbers (six numbers) and order them depending on the X axis positions.
- save the detection to a MongoDB.
- snap a photo of the plate.
- compress the photo.
- encode the photo as a string and save it to the cloud.
All these results are also connected to a FULTTER app for monitoring the live stream, and cheching the results.