# Recognising text characters from an image

**URL:** <https://forum.kirupa.com/t/recognising-text-characters-from-an-image/328228>\
**Category:** flash\
**Created:** [April 30, 2012, 3:50am UTC](https://forum.kirupa.com/t/recognising-text-characters-from-an-image/328228 "2012-04-30T03:50:21Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![four4swords](https://avatars.discourse-cdn.com/v4/letter/f/87869e/32.png) [@four4swords](https://forum.kirupa.com/u/four4swords)\
**Post date:** [April 30, 2012, 3:50am UTC](https://forum.kirupa.com/t/recognising-text-characters-from-an-image/328228/1 "2012-04-30T03:50:21Z")

</div>

Hi,

So let’s say you have an image of a letter, for example, let’s use the popular Words with Friends letters:

![](http://forum.kirupa.com/uploads/kirupa/6625/296016693dc841b1.png)

What would be the best way to detect each letter from this image?

At first I collected images of all the black letters and saved them into a database, then use that to compare the 2 images, if they are sufficiently similar, it would then be assigned the corresponding letter.

But my way has many limitations, such as if a letter with a different coloured font is used, then the comparison would not work.

What better options are there for detecting the letters?