ZERO OCR backend

purpose

This is the core of the project OCR. By this I want to gain proficiency in the following aspects.

• Convert given image (scanned image using NN) to text.
• Log the requests for errors, debugging, analytics(Logging).
• Write tests to make sure of the speed and delivery for customers(Testing).

Logging

Log every request to csv so that it easy import with pandas and visualize. Let there be a request made with an image.

• Log info Time-IPaddress-ImageSize-ResponseTime-Status(200,400,500).
• Log error if status is 400 or 500 and save image with Time's epoch seconds.

Testing

• Tests before deployment.
  • Can deliver a test image with 200.
• Tests for performance.
  • Check number of parallel connections supported
  • Check number of requests per minute

deployed at zerocr.herokuapp.com

frontend code: ocr_frontned