Extraction Of Tabular Data From PDFs Using Python

Pic source : Google

How using python?

We can extract tabular data from PDFs using camelot library in python with >90% accuracy and we can save into csv or excel file.

What is camelot?

Camelot is python based,MIT licensed ,open source library having following features:

• Work well and configurable
• We can debug and visualize using python matplotlib library
• We can export output file as a csv or excel file
• Camelot have excellent documentation

Installation :

Using Conda:

• conda install camelot-py -c conda forge

Using pip (after installing tk and ghostscript)

• pip install camelot-py[cv]

Note : It only works with text based PDFs not scanned documents.

Others PDFs Extraction Tools Available:

• Tabula- Java based,Open source
• pdfplumber- Python,Opensource
• pdftables- Python,proprietary and paid
• Smallpdfs- Online and paid service

Problems with these solutions:

• We can not save output file as csv or excel.
• These tools are not scalable and maintainable.

Conclusion:

This article is inspired by speaker Vinayak Mehta in PyconIndia 2019.Thank you for reading. Please give it a try, have fun and let me know your feedback!