Microsoft AI School - AI Computer Vision in Azure

A Simple Introduction to Computer Vision

Computer Vision is a field of AI that renders machines the power to emulate human vision to perceive the world visually by processing images, either from a live camera feed or from digital photographs or videos. Some exciting examples of computer vision in practice today are: autonomous vehicles, google translate app, facial recognition, healthcare, real-time sports tracking, agriculture, and manufacturing. 

There are numerous types of computer vision problems that data scientists and AI engineers can solve with the help of a mix of custom machine learning models and platform-as-a-service (PaaS) solutions. Microsoft Azure offers you numerous cognitive services that have prebuilt computer vision model capabilities. 

Learn AI Computer Vision with Azure

Microsoft AI School hosts an impressive list of modules to help you get started with Computer Vision in Azure. To conquer Computer Vision in Azure, you may follow the following learning path featured in Microsoft’s AI School - “Explore computer vision in Microsoft Azure.” This learning path has six modules. Read ahead to get a quick peek at each module of this learning path. 

Prerequisite: You must be able to navigate the Azure portal to pursue this learning path.

Module1: Analyze Images with The Computer Vision Service

Computer Vision is a cognitive service in Microsoft Azure that analyzes content in images and videos and returns detailed information about an image and the objects depicted by the image.

In the first module, you will learn how to use the computer vision service in Azure to analyze images. The primary objectives of this module are to help you:

  • Identify image analysis tasks that can be performed with Azure’s Computer Vision service.
  • Provision a Computer Vision resource.
  • Use a Computer Vision resource to analyze an image.

Here’s an overview of the units covered within this module:

Module 2: Classify Images with The Custom Vision service

Image classification is the field of Artificial Intelligence that harnesses the predictive capabilities of a machine learning model to enable AI systems to detect objects or real-world items within an image based on their unique properties. Some potential use cases of image classification are: product identification, disaster investigation, medical diagnosis, and anomaly detection. Microsoft Azure’s Custom Vision is a cognitive service in Azure that encapsulates common techniques to train and publish image classification models as a software service. This allows you to deploy classification models with minimal knowledge of deep learning techniques. 

Thus, this module will help you learn how to use the Computer Vision service in Azure to create an image classification model. The primary objectives of this module are to help you:

  • Identify image classification scenarios and technologies.
  • Provision a Custom Vision resource and use the Custom Vision portal.
  • Train an image classification model.
  • Publish and consume an image classification model.

Here’s an overview of the units covered within this module:

Module 3: Detect Objects in Images with The Custom Vision Service

Object detection is a computer vision task in which a model is trained to recognize individual objects in an image and identify their location in the image. Some common applications of object detection are: evaluating building safety, creating softwares for self-driving cars, medical imaging such as MRI and x-rays for diagnosis.

Note: Image Classification is used to categorize images based on the primary subject. Object detection goes one step further and identifies individual objects within the image and returns the coordinates of a bounding box that indicates the object's location. 

Thus, in the third module of this learning path, you will learn to use the Custom Vision service in Azure to create an object detection solution. 

Here’s an overview of the units covered within this module:

Module 4: Detect and Analyze Faces with The Face Service

Face detection, analysis, and recognition is an area of artificial intelligence that allows you to use algorithms to analyze and locate human faces in images or videos. Azure Face is a cognitive service in Azure that consists of AI algorithms that can detect, recognize, and analyze human faces in images.

Thus, the objective of this module is to teach you with a hands-on example- “How to use the Face the cognitive service to analyze and detect images in images.” 

Here’s an overview of the units covered within this module:

Module 5: Read text with the Computer Vision service

Optical Character Recognition (OCR), also known as Text Recognition, allows AI systems to identify characters like letters, numbers and texts within an image. This enables applications to extract information from photographs, scanned documents, and other sources of digitized text.

The objective of this module is to help you learn - “how to read text in images with the Computer Vision service.”

Here’s an overview of the units covered within this module:

Module 6: Analyze receipts with the Form Recognizer service

Form Recognizer in Azure is an AI-powered document extraction service that can be used to automate the processing of data in documents such as forms, invoices, and receipts. It combines OCR with predictive models that can interpret form data. Processing invoices and forms is a day-to-day task in most organizations who are looking to automate the process of data extraction from receipts on a regular basis. Thus, the final module of this learning path teaches you how to use the built-in receipt processing capabilities of the Form Recognizer service in Azure. 

Here’s an overview of the units covered within this module:


In summary, if you are looking to master the concepts of computer vision with the help of real-time examples and exercises, then this learning path will serve as the perfect roadmap for you. You will learn how to implement numerous cognitive services in Azure and build solutions without requiring in-depth knowledge of deep learning and complex AI algorithms. 

Happy vision coding!