sovse/tflite_avto_num_recognation
License plate recognition . Model training and conversion to tflite
Implements a complete pipeline combining object detection (ResNet50-based plate localization), geometric alignment, and character-level OCR using TensorFlow 2/Keras for training with TensorFlow Lite inference optimization. Targets Russian license plate formats through a multi-stage approach: detecting and cropping plates, aligning skewed images, then recognizing individual characters. Designed for deployment on edge devices (embedded controllers, mobile phones) via TFLite quantized models.
No commits in the last 6 months.
Stars
42
Forks
10
Language
Jupyter Notebook
License
—
Category
Last pushed
Mar 11, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/sovse/tflite_avto_num_recognation"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
awslabs/aws-ai-solution-kit
Machine Learning APIs for common use cases, include: General OCR (Simplified/Traditional...
Dharun/Tensorflow-License-Plate-Detection
The project developed using TensorFlow to recognize the License Plate from a car and to detect...
ValerioSpagnoli/EyeRoad
Vehicles and plate detection and tracking with string plate recognition.
pharmbio/plaid
PLAID (Plate Layouts using Artificial Intelligence Design) is a flexible constraint-programming...
santifiorino/license-plate-recognition
Argentinian License Plate Recognizer