https://github.com/preetraj2002/tablify

Converts a snapshot of a table (an image) into tabular data using OCR. It using image processing and enhancement techniques to help with the OCR.
https://github.com/preetraj2002/tablify

ocr opencv otsu-thresholding pytesseract pytesseract-ocr python tabular-data

Last synced: 2 months ago
JSON representation

Converts a snapshot of a table (an image) into tabular data using OCR. It using image processing and enhancement techniques to help with the OCR.

• Host: GitHub
• URL: https://github.com/preetraj2002/tablify
• Owner: Preetraj2002
• Created: 2024-12-05T18:22:22.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-12-05T19:38:15.000Z (over 1 year ago)
• Last Synced: 2025-04-09T00:07:20.021Z (about 1 year ago)
• Topics: ocr, opencv, otsu-thresholding, pytesseract, pytesseract-ocr, python, tabular-data
• Language: Python
• Homepage:
• Size: 872 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# **Tablify** - Convert Images to Tabular Data using OCR

**Tablify** is a Python-based tool that converts tabular data from images into CSV files using Optical Character Recognition (OCR). It processes images, extracts the text using `pytesseract`, and organizes it into rows and columns for easy data extraction and analysis.

## **Features:**

- Converts images of tables into structured CSV files.

- Uses `pytesseract` to perform OCR on images.

- Processes images to detect individual text blocks, sort them by coordinates, and group them into rows.

## **Installation:**

1. **Clone the repository:**

   ```bash

   git clone https://github.com/Preetraj2002/Tablify.git

   cd Tablify

   ```

2. **Install required dependencies:**

   Make sure you have Python 3.x installed. Then, install the required libraries:

   ```bash

   pip install -r requirements.txt

   ```

3. **Install Tesseract OCR:**

   - **Windows:** Download the Tesseract installer from [here](https://github.com/UB-Mannheim/tesseract/wiki) and add the path to your system environment variables.

   - **Linux:** Install Tesseract using:

     ```bash

     sudo apt install tesseract-ocr

     ```

   - **macOS:** Use Homebrew to install Tesseract:

     ```bash

     brew install tesseract

     ```

## **How to Use:**

1. **Prepare an Image:**

   Ensure the image contains tabular data that you want to extract. The tool works best with clear, well-contrasted images.

2. **Run the Script:**

   After setting up, simply run the script on your image:

   ```bash

   python tablify.py path/to/your/image.jpg

   ```

   This will generate a `output.csv` file in the same directory.

3. **Check the Output:**

   Open `output.csv` to see the extracted table data in tabular format.

## **Process Inside Tablify:**

1. **Image Preprocessing:**

   The image is converted to grayscale, and a binary thresholding is applied to make the text clearer for OCR.

   Original:

   ![Original](images/image_csv.jpeg)

   Grayscale:

   ![Gray](images/gray_image.png)

   After OTSU thresholding:

   ![Thresholded_image](images/thresholded_image.png)

   Dilation:

   ![Dilation](images/dilation.png)

   

2. **Contour Detection:**

   Using OpenCV, contours of the text blocks are identified to group text into rows and columns.

   Marked Countours:

   ![Countours](images/countours.png)

   Marked Centroids of the countours:

   ![centroids](images/centroids_with_labels.png)

   

3. **Text Extraction:**

   Each text block is processed with `pytesseract` to extract the text, which is then organized into a structured CSV format.

4. **CSV Generation:**

   The processed text is organized into rows based on vertical alignment and saved as a CSV file.

## **License:**

This project is licensed under the MIT License