

پروژه ایجاد سیستم تشخیص پلاک ایرانی با زبان c++ با مشخصات زیر We are looking for an experienced C++ developer with strong expertise in computer vision and video streaming to build a real-time license plate detection and recognition system. The system should be containerized using Docker, consume an RTSP video stream from IP cameras, detect vehicle license plates, and extract plate numbers using OCR.
This solution will be used in a production environment, so it must be efficient, stable, and easy to deploy.
Project Requirements:
Core Features: RTSP Video Stream Input
Configurable via environment variable or config file
Use GStreamer for robust and low-latency video handling
License Plate Detection
Implement or integrate a YOLO-based model (YOLOv5/v8 or YOLOv4-tiny)
Detect license plates in real-time (at least 5 FPS)
Plate Recognition (OCR)
Extract and recognize text from detected plates
Use Tesseract OCR or a deep learning-based CRNN model
Dockerized Deployment
Package the application in a Docker container
Ensure lightweight and clean runtime environment
Output Handling
Print detected plate numbers to console and/or log to file
Optionally, expose results through a simple REST API or send to an external service
Technical Expectations:
Language: C++
Libraries: OpenCV, GStreamer, Tesseract or custom OCR
Optional: CUDA/NVIDIA GPU acceleration support
Well-documented code and Docker setup
Clean, modular, and extensible architecture
Deliverables:
Full source code (C++) with documentation
Dockerfile and build instructions
Sample config.json or .env file for configuration
README with usage instructions
(Optional) Demo video or screenshots
Bonus Skills (Nice to Have):
Experience with TensorRT or YOLOv8 ONNX models
Experience with edge devices (e.g., NVIDIA Jetson)
Prior work in ALPR (Automatic License Plate Recognition) projects
To Apply, Please Include:
A brief proposal explaining your approach
Relevant experience or sample projects
Time estimate and cost
Any suggestions to improve the system


