https://github.com/negarhonarvar/equation-solving-ocr

An Equation Solving Model for hand written equations, enahned by Tesseract
https://github.com/negarhonarvar/equation-solving-ocr

cv2 pil pytorch tesseract-ocr torchvision

Last synced: 6 months ago
JSON representation

An Equation Solving Model for hand written equations, enahned by Tesseract

• Host: GitHub
• URL: https://github.com/negarhonarvar/equation-solving-ocr
• Owner: negarhonarvar
• License: mit
• Created: 2025-04-12T20:22:06.000Z (6 months ago)
• Default Branch: main
• Last Pushed: 2025-04-12T21:58:14.000Z (6 months ago)
• Last Synced: 2025-04-13T11:14:27.738Z (6 months ago)
• Topics: cv2, pil, pytorch, tesseract-ocr, torchvision
• Language: Jupyter Notebook
• Homepage:
• Size: 45.9 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Equation Solver with OCR

This repository implements a full pipeline to classify and solve equations from images using deep learning and OCR techniques. It supports both **handwritten** and **typed** expressions and reconstructs them into solvable strings.

## Overview 🚀

The system processes input equation images by:

- Classifying the equation as handwritten or typed

- Segmenting and recognizing characters using:

  - CNN trained on MNIST and a custom symbol dataset

  - Tesseract and EasyOCR for OCR-based recognition

- Reconstructing the equation

- Solving the expression and saving results

## Dataset 📁

All training and test data are available in this shared Google Drive folder:

[Equation Dataset - Google Drive](https://drive.google.com/drive/folders/13nICjIQg57NhPn5eYJwZp1jCpJwk4vjp)

## How to Use 🛠

### 1. Clone the Repository

    git clone https://github.com/yourusername/equation-ocr-solver.git

    cd equation-ocr-solver

    

### 2. Install Dependencies

    pip install -r requirements.txt

    

If you're running in Google Colab, also install Tesseract with:

    !apt-get install tesseract-ocr

    

### 3. Run the Notebook

Open and run the following notebook in Google Colab:

    EquationSolving-OCR.ipynb

### 4. Output

After running the full pipeline, a file named submission.csv will be generated containing:

  - The predicted type of each equation image (typed or handwritten)

  - The evaluated result of the expression

## Project Structure 📦

    ├── README.md

    ├── requirements.txt

    ├── EquationSolving-OCR.ipynb         # Original notebook (Colab)

    ├── equationsolving_ocr.py            # Exported script version of the notebook

    ├── drive/

    │   ├── train/                         # Training images

    │   ├── test/                          # Test images

    │   ├── train_info.csv                 # Labels for training

    │   └── submission.csv                 # Generated predictions

## License 🧾

This project is licensed under the MIT License. See the LICENSE file for details.