How to detect handwriting with Azure Computer Vision and Azure Vision Studio

Published Aug 04 2022 10:09 AM 523 Views
Microsoft

bethanyjep_0-1659533736113.png

 

Have you ever wanted to build and explore the different features of Azure Computer Vision? The Azure Vision Studio enables you to experiment with the different features and is now available in public preview. 

Introduction 

Computer vision is one of the core areas of artificial intelligence and can enable your solution to ‘see’ images and videos and make sense of them. Azure Computer Vision Service is a prebuilt computer vision solution that allows you to analyze images, recognize text and detect objects in images without writing a single line of code. 

What can you do with Vision Studio? 

Using the vision studio, you can test out different Computer Vision services including Face Identification, Image Tagging, Image Captioning, and many more. The UX is rich with pointers on Responsible AI as well as a guide on additional resources. 

bethanyjep_1-1659533736115.png 

Let’s get started with Vision Studio. 

Step 1: log in to Vision Studio (azure.com) with your school or work email that is linked to an Azure account. If you do not have a work or school email, follow instructions at the end of the blog on how you can use your personal email. 

bethanyjep_2-1659533736117.png

Step 2: Select the feature you would like to test out. For this demo, we will go with “Extract text from images” feature. Using the feature, you can detect text from either printed or handwritten documents. 

bethanyjep_3-1659533736119.png

Step 3: The demo will utilize your Azure resources and some costs will be incurred. Therefore, you first need to accept the terms. 

bethanyjep_4-1659533736120.png

Step 3: Once you acknowledge the terms, go ahead and either select a pre-existing resource or create a new cognitive service resource. 

bethanyjep_5-1659533736121.png

Step 4: Time to test it out. Upload or take a photo with your device and test to see the functionality. In my case, I scribbled some notes in my notebook, and here are the results. 

bethanyjep_6-1659533736127.png

Now it is your turn, try it out 

Vision Studio at Vision Studio (azure.com), you can test it out with your work or school email. 

For Personal accounts, you will have to create a new user to test it out. Follow the steps below: 

What next? 

Watch this short clip to see the demo in action.

You can now integrate Optical Character Recognition (OCR) with your application. Read more on how to use the REST API or SDK QuickStart to intergrate the features. 

Co-Authors
Version history
Last update:
‎Aug 04 2022 10:09 AM
Updated by: