Artificial Intelligence

Introducing Microsoft AI School: Extracting Information from Images and Documents

Welcome to Microsoft AI School! Are you interested in learning how to extract valuable information from images and documents using advanced AI tools and services? Look no further than our comprehensive learning path, which will guide you through the process step by step. To get started, click on this link: “Extract text from images and documents.”

Before jumping into the course, there are a few prerequisites you need to fulfill. You must have an active Azure account, be able to navigate the Azure portal, and have some programming knowledge in either Python or C#. Once you have met these requirements, you are ready to embark on this exciting learning journey.

Our learning path consists of two modules, each covering essential aspects of extracting information from images and documents. Let’s take a quick look at what you will learn in each module.

In today’s digital age, companies across various industries often deal with large volumes of images containing embedded texts. The task of extracting and storing this information in a database would be exceptionally challenging without the assistance of AI services. With the aid of AI services like Microsoft Azure’s Computer Vision, complex scenarios involving text extraction from images can be solved effortlessly using pre-defined algorithms.

In the first module of our learning path, you will explore how to:

  • Utilize Azure Computer Vision service through SDKs and REST API.
  • Learn about the OCR API and Read API, which allow computer vision services to read texts from images.
  • Develop an application capable of reading both handwritten and printed texts.

Computer Vision is a powerful AI service offered by Azure. It can analyze image and video content, providing valuable insights. When it comes to reading texts from images, Computer Vision offers two APIs: the OCR API and the Read API. The OCR API can read texts in multiple languages from small to medium volumes of images, while the Read API delivers even higher accuracy and can handle small to large volumes of texts.

In the first few units of this module, you will learn how to access these APIs using the REST API client library and obtain JSON responses. You will then put your newfound knowledge into practice through a hands-on exercise.

Every industry relies on forms to collect and communicate vital information. However, manually extracting and transcribing data from thousands of forms can be an overwhelming task. This is where Azure Form Recognizer comes to the rescue.

Form Recognizer is a cognitive service that uses REST APIs and client library SDKs to extract data from forms with high levels of accuracy. By leveraging Optical Character Recognition (OCR) and deep learning models, Form Recognizer can efficiently extract key-value pairs and table data from various documents. It consists of the following services:

  • Layout Service: This service accepts various file formats, such as JPEG, PNG, PDF, and TIFF, and returns a JSON file with text location, tables, and other key information.
  • Prebuilt Models: Form Recognizer supports pre-built models for extracting information from receipts, business cards, and invoices.
  • Custom Models: You can also train custom models specific to your business needs using supervised or unsupervised learning.

The Form Recognizer service has numerous applications, including process automation, knowledge mining, and industry-specific solutions.

To integrate Form Recognizer services into your applications or workflows, you can use the REST API or client library SDKs. There is also a user interface called the Form OCR Test Tool (FOTT), which enables layout extraction and model training.

The objective of the second module is to teach you how to:

  • Automate processes using FormRecognizer’s Layout service, prebuilt models, and custom service.
  • Utilize OCR capabilities with SDKs, REST API, and the Form OCR Test Tool (FOTT).
  • Create and test custom models using supervised and unsupervised training.


If you or your organization faces the challenge of extracting vast amounts of data from images and documents, Microsoft Azure’s Computer Vision capabilities offer the ideal solution for automating this task. To master the art of data extraction using computer vision in Azure, follow this link: It will provide you with valuable insights and equip you with the skills needed to excel in this field.

At Skrots, we also offer similar services to help you extract information from images and documents effortlessly. Visit to explore our range of services and learn how we can assist you in your data extraction journey. Check out all the services we provide at Thank you for choosing Skrots!

Happy AI coding!

Show More

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button