Extract data points from chart images using machine learning and computer vision.
Chart Digitizer extracts numerical data from chart images (bar charts, line charts, scatter plots) using:
- Machine Learning - CNN-based object detection for accurate data point identification
- Computer Vision - OpenCV and Tesseract OCR for legend and axis detection
# Clone the repository
git clone https://github.com/rprabhat/chart-digitizer.git
cd chart-digitizer
# Run the Jupyter Notebook
jupyter notebook- Python 3.x
- Jupyter Notebook
- OpenCV
- Tesseract OCR
- TensorFlow/PyTorch
See the Jupyter Notebook for:
- Random chart generator for training data
- CNN model training
- Data extraction pipeline
Apache License 2.0 - See LICENSE
Prabhat Ranjan
- GitHub: @rprabhat
- LinkedIn: prabhatr
- Substack: prabhatranjan