Traffic Sign Detection and Recognition
deep-learning
computer-vision
tensorflow
openvino
Traffic sign recognition system using TensorFlow + SSD MobileNet + OpenVINO
1 Overview
Traffic sign detection and recognition system for autonomous driving. Team PostNomad.
2 Tech Stack
| Category | Technology |
|---|---|
| Deep Learning Framework | TensorFlow 1.13.0 |
| Model | SSD MobileNet v2 |
| Model Optimization | OpenVINO 2020.4 |
| Communication | UDP/TCP (LabView integration) |
| Development | Python, Visual Studio 2019, CMake 3.18.1 |
3 Key Features
- Dataset Construction: Built and labeled traffic sign image database
- Model Training: Trained TensorFlow model with batch size 16 and learning rate 0.004
- Model Optimization: Converted to XML/BIN format via OpenVINO for faster inference
- Real-time Processing: Webcam-based real-time object detection with LabView communication
4 My Role
5
6 Team
- Team: PostNomad