In document recording, there are many times when you receive pdf documents from many different types of documents in pdf documents. And for humans, what they do is they will review page by page and then use a pdf editor to break pages and files into different folders. In this article we will discuss how to sort documents using image recognition

If you use a robot to do this, can it be done? Yes Because we use image recognition to identify the document.

So the first time you do that is, you need to train an image recognition engine like Gleematic to recognize the layout of the document or part of the document like the letter header.

After that is complete, when we receive the pdf document, we need to split the pdf document into each pdf with 1 page. Then we need to convert pdf to images and make an introduction to images.

Then, we will know which part of the page belongs to a different type of document. So we can then use folders and file movements to move documents to these different folders.

The main problem why Gleematic can make this happen is, that it comes by doing it yourself without the machine code drawing training. And it can also split large pdf files into pages and move documents to folders.

Its main advantages are:

  • Reducing human power to do tedious tasks
  • Increase processing speed
  • Reducing lap times

What about documents with missing signatures or stamps?

By using image recognition, we can overcome it too. This can free humans from doing the tedious tasks of checking missing signatures and stamps. Let cognitive automation do the work to examine our work.

To check if a signature is missing, we allow image recognition to learn what a blank signature looks like. Because Gleematic is trained to recognize that.

Written by: Christopher Lim