TraceVision: Real-Time Industrial Logging & Vision Intelligence TraceVision is an end-to-end monitoring solution designed to bridge the gap between raw computer vision and actionable operational data. By integrating a real-time detection pipeline with a scalable logging backend, TraceVision enables high-fidelity tracking of objects and events, providing a foundation for bottleneck analysis and automated auditing in manufacturing and logistics environments.

Technical Architecture The system is built on a decoupled, high-performance stack designed for low-latency processing and persistent data integrity:

Vision Engine (Python/Flask): Utilizes a specialized detection pipeline to process live webcam streams, performing real-time inference and frame-by-frame analysis.

Orchestration Layer (Node.js/Express): Acts as the central nervous system, managing state, handling API routing, and interfacing with the database.

Persistent Storage (MongoDB): A NoSQL approach to logging, storing structured event data including timestamps, object classifications, and confidence scores.

Cloud Infrastructure (Cloudinary): Offloads heavy image assets to the cloud, ensuring that every detection event is backed by a visual record without taxing local storage.

Interface (React): A responsive dashboard that provides a live telemetry feed and a searchable history of logged production events.

Installation & Deployment

1. Repository Initialization Bash git clone https://github.com/manchxz/TraceVision.git cd TraceVision
2. Vision Engine Setup (Flask) It is recommended to use a virtual environment to isolate dependencies.

Bash cd backend-python python -m venv venv

Windows: venv\Scripts\activate | Mac/Linux: source venv/bin/activate

pip install -r requirements.txt python app.py 3. Middleware & Logic Setup (Express) Bash cd backend-node npm install npm run server 4. Frontend Dashboard (React) Bash cd frontend npm install npm run dev Configuration (Environment Variables) Create a .env file in the root of the backend directory. The system requires these keys to bridge the vision engine with the cloud storage and database:

Code snippet MONGODB_URI = "your_mongodb_connection_string" JWT_SECRET = "your_secure_hash_key"

CLOUDINARY_NAME = "your_cloud_name" CLOUDINARY_API_KEY = "your_api_key" CLOUDINARY_SECRET_KEY = "your_api_secret" Roadmap & Future Optimizations Edge Optimization: Implementing model quantization to allow the Vision Engine to run on low-power industrial edge devices (Raspberry Pi/NVIDIA Jetson).

Predictive Analytics: Integrating an XGBoost-based layer to predict production delays based on real-time logging frequency.

Advanced Trajectory Mapping: Transitioning from simple detection to movement-path analysis for worker-efficiency benchmarking.

Credits & Collaboration Shoaib Imran: Core Vision Architecture & Tracking Logic.

Manish Mahto: Industrial Logic Integration, Operational Benchmarking, and Optimization for Manufacturing Constraints.