What is OCR?

Optical character Recognition, OCR, is a computer software which enables the recognition of characters, words and phrases in a digital format from scanned, handwritten and printed paper documents, and even objects within pictures.

An OCR engine has the capability to structure the data it extracts and classify documents. This allows the document to be digitally manipulated, permitting it to be edited and searched for mistakes that were previously missed by the original document author.

How does OCR work?

An OCR Engine is a machine that continuously learns while never forgetting previously learnt information. A brand-new OCR would have to be fed data. The second time is when the OCR will recognise the information and be able to structure it.
The following steps are taken by the OCR:
  1. The paper document is scanned through the OCR Engine
  2. The Raster image is then converted into a digital document by the OCR Software
  3. Scanned document is analysed for light and dark areas to differentiate between characters
  4. Converted into ASCII Code allowing the software to recognise the numbers and letters.
  5. The document can then be saved in whatever format you want (e.g. DOC, PDF, HTML) or can be exported onto Microsoft Office or Google Docs
However, if you want to be able to edit the text on a brochure, for instance, you could use OCR software to detect the letters and numbers in the scanned document and eventually those letters will be put into words and sentences, creating a digital document. Therefore, you are now able to edit what you want in the brochure

Why use OCR?

The main advantages of using OCR software is that it can easily distinguish between various fonts, so most writing can be read. The program is very fast at reading the documents reliably and can even understand the text irrespective of its rotation and inversion.

Also, lots of time is saved if searching for an answer in book or piece of text, as once scanned, the identification of the solution is far quicker and simpler.

As you can see, OCR plays an important role for several different reasons which opens many more doors when it comes to your document-based information. The technology has also shown its immense use for helping digitalise historic files, old texts and newspapers which makes them more easily accessible to the public.

Other uses of OCR, other than allowing paper documents to be edited, are:
  • Used in airports at passport controls for passports and other related documents.
  • Used by banks for invoices and bank statements
  • Business cards
  • Voucher and promotional code scanning
  • Ticket and ID validation
  • It plays a key role in Trade Finance (see below).


The whole recognition process is far quicker than a human and therefore saves a lot of time checking for mistakes. This is what we do here at Traydstream.

The current problem we face is that trade is done manually, as in documents are being checked by people to make sure that the hundreds of thousands of rules are being followed. Obviously, this process is monotonous and long with transactions taking almost a week to occur and nearly two-thirds of seller's documents fail to meet Letter of Credit (LC) terms and conditions.

Traydsteam has noticed this problem in international trade finance and have found a solution. Through automation and machine learning, we are now able to improve efficiency of trade and reduce operating costs by over 50%.

So, the role of OCR in Traydstream is to scan the paper documents and convert them to digital information. The good news is that there is no requirement for any significant hardware investment. This digitised information is then analysed using automation.

Documents can now be checked with the rules from the UCP 600 and the ISBP. Thus, OCR enables documents and Letters of Credit to be checked to a higher accuracy and speeds up the process.

Other Links
Scroll to top