gagangulyani / covid-text-extractor

OCR made for the specific use case of extracting Covid Info from Images, PDFs and Texts

Welcome to my Profile! 😎

About Me:

I am a software engineer and a music producer. I have experience working with the following technologies:

• Python
• Flask
• JavaScript
• NodeJS
• jQuery
• RegEx
• Google APIs
• MongoDB
• SQLite3
• MySQL
• Redis
• git
• GitHub
• Bash
• OpenCV

Where to find me:

• Twitter
• Instagram (for music)
• Email Me

covid-text-extractor's Issues

Add Documentation

The functions written in the current project aren't documented so it might be difficult for someone who never worked with it to take more time to understand how to code works.

So, Documenting the code is much needed.

Using Docker for Containerization

Text Parsing for phone numbers on PDF sometimes gives invalid numbers

• Text Parsing for PDFs get some invalid Phone Numbers (merging two different numbers or pin codes as one)

Regex Invalid for Some Phone Numbers

To-Do:

• Regular Expression is invalid for some phone numbers

Refactoring the Code base

The Current Project needs to be refactored and restructured in order to make it cleaner, readable and more Pythonic.

Using Fast API for building API

The main goal for this project is to create a free API that can be used by Bots and Web Apps.

The fastest way to build an API is using FastAPI.

Deploy it on a Platform

After building everything, Deploy it.

Create Unit Tests for the Extractor

Test cases need to be made as soon as possible to make sure that new changes don't affect the existing code negatively.

To-Do:

According to its required attributes (numbers etc)...

• Make Unit Tests for Each Image
• Make Unit Tests for PDF files

Creating a Web App for the API

Creating a User Interface for giving access to the API is needed.

Optimize Algorithms

To-Do:

For Optimizing Algorithms

• Optimize Image Preprocessing
• Optimize Text Parser