OCR Blocks in PictoBlox

The Artificial Intelligence extension in PictoBlox has blocks dedicated to OCR. Let’s first add the extension to our project:

Create a new project in PictoBlox.
Next, click the Add Extension button and add the Artificial Intelligence extension.

Now, let’s have a look at the blocks:

Text Recognition Blocks

For text recognition, we’ll use the same blocks that we used in computer vision. You have to select the type of text you want to recognize in the image. The two types of text that the block can recognize are:

Handwritten text
Printed text

You can input the image in the following ways:

Camera feed
URL – online link of the image
From the stage, backdrop, or costume

Recognizing Printed Text

Once you have analyzed the image for printed text, you can use a printed text result block to get the recognized text from the image in a string format.

E.g., if you input the image above in one of the recognize blocks, the recognized text will be as follows:

Recognizing Handwritten Text

Once you have analyzed the images for handwritten text, you can use a handwritten text result block to get the recognized text from the image in a string format.

E.g., if you input the image above in one of the recognize blocks, the recognized text will be as follows. You can extract the city and Pincode easily with simple string manipulation.

Limitations of OCR

The accuracy of text recognition operations depends on the quality of the images. The following factors may lead to an inaccurate reading:

Blurry images
Handwritten or cursive text
Artistic font styles
Small text size
Complex backgrounds, shadows, glare over text, or perspective distortion
Oversized or missing capital letters at the beginnings of words
Subscript, superscript, or strikethrough text

In the next, topic we’ll use these blocks to make a fun PictoBlox project!