Identifying What OCR or Optical Character Recognition is and Its Functions and how it Works


As the years went on, various daily activities are also demanded to be faster, more efficient, and flexible. To meet these demands, new ways to do these activities continue to emerge while supported by increasingly advanced technology, thus making difficult to simple tasks very easy to do. Many business companies need to digitize their documents to be stored in a central repository to make them more unified and structured. Therefore, character recognition is a popular method to digitize these documents.

CR or character recognition is a process that allows a computer to recognize text writings or printed texts in the form of numbers or letters before converting it into a form that is readable by a computer. In other words, CR converts an image into a text.

Initially, this technology was developed in the early 20th century to help the blind people in reading texts. In this era, Emmanuel Goldberg invented a device known as written text decoder. At the same time, Edmund Fournier d’Albe developed a tool, namely the Optophone, which was a handheld scanner that can digitize letters printed in text or documents. Then, this technology was developed specifically for the visually impaired where this tool can read text with vocals to the user. With this discovery, they are considered as pioneers of CR technology.

Then, how does CR work? One of the most frequently used methods is OCR (Optical Character Recognition). OCR is a tool or software used to recognize characters from an image and convert them into a digital text format file by a computer automatically, thus making it easier to be processed.

The results of the OCR can be used for various purposes, such as editing the document, quoting the text, and even loading it on the website without having to type the text manually. Judging from its function, OCR is also widely used in other applications such as translator applications, text reviewer applications, and many others. This technology is also able to help people with special needs in understanding text on a computer screen, by integrating it with text-to-speech applications. Moreover, by adding AI (Artificial Intelligence), this OCR technology is capable in recognizing various forms of writing to the languages used in these texts.

Generally, OCR’s work process is as follows:

a. File Input

The inputted files are files with the *.bmp or *.jpg formats.

b. Pre-processing

Pre-processing is a process in omitting unnecessary parts in the input image before moving to the next stage.

c. Segmentation

Segmentation is a process of distinguishing surveillance areas in each detected letter character.

d. Normalization

Normalization is a process of changing each character’s regional dimension and density.

e. Extraction

Extraction is a process of attaining certain characteristics from the observed characters.

f. Recognition

Recognition is a process of recognizing observed characters by comparing the characters’ characteristics with the similar characteristics in a certain database. Now that we know what character recognition is, then how can we choose the best OCR software? Choosing OCR software is very important as it can determine the success of your business project. By choosing the right vendor according to your needs, you can avoid problems that can hinder your work. Observe the software’s performance, flexibility, ease of use, and reliability before buying the software. AdIns also takes advantage of this OCR technology to simplify the process of digitizing documents that are integrated in one of the AdIns services, namely Lite DMS. Lite DMS is a document management system specifically designed to structure documents and simplify the document search process.


Sofani, Rachmah, 2009, “Sistem OCR”,   http://digilib.ittelkom.ac.id/index.php?option=com_content&view=article&id=413:sistem-ocr&catid=15:pemrosesan-sinyal&Itemid=14.

Photo source: