VisionAI is a modern, high-performance web application designed to detect image quality. Using advanced Machine Learning and Computer Vision techniques, it accurately classifies images as either Sharp or Blurred.
- Modern Dark UI: Iconic glassmorphism design with a seamless user experience.
- MLP Classifier: Powered by a Multi-Layer Perceptron model trained on extracted image features.
- Real-time Analysis: Instant feedback on image clarity with confidence scoring.
- Responsive Design: Optimized for both desktop and mobile viewing.
- Frontend: HTML5, Tailwind CSS (Modern Glass UI)
- Backend: Python, Flask
- Machine Learning: Scikit-learn (MLP Classifier), Joblib
- Computer Vision: OpenCV (Sobel, Tenengrad Feature Extraction)
- Deployment: Vercel (Serverless Functions)
The application uses Feature Engineering to analyze the input image:
- Sobel Variance: Measures the intensity of edges.
- Tenengrad Mean: Evaluates the focus level of the image.
- ML Prediction: These features are fed into a pre-trained MLP model to determine the final class.
- Python 3.9+
- A Vercel account for deployment.
- Clone the repository:
git clone https://github.com
- Install dependencies:
pip install -r requirements.txt - Run locally :
python api/index.py
Live Demo: View VisionAI App
If you find this project useful or like the UI, feel free to drop a ⭐ on this repository. It means a lot!