In this course, you'll learn about Amazon Rekognition, a service that enables you to easily and quickly integrate computer vision features directly into your own applications. At its core, Amazon Rekognition provides an API that you submit images and/or videos to. You then instruct the Rekognition service to perform a specific analysis on the media. The analysis can be anything from detecting faces within an image to extracting labels from a video in an asynchronous manner.
Lectures
We’ll provide in-depth reviews of the image, video, and collection based API sets. Amazon Rekognition makes it easy to add image and video analysis to your applications. You simply submit your image or video to the Rekognition service, and the service will then identify objects, people, text, scenes, emotions, and activities. Additionally the service can be used to moderate content, by detecting inappropriate or objectionable content.
Demonstrations
We finish the course with a couple of demonstrations:
- In the 1st demonstration we will show you how to use Amazon Rekognition together with S3 and CloudFront for hosting, to build a simple facial analysis static web application.
- In the 2nd demonstration we will show you how to use the AWSCLI to implement an object and feature detection system that sends an email when certain features are detected within an image.
Both demonstrations will highlight the capabilities of the Amazon Rekognition and the different approaches you can adopt to interface with the service.
Intended Audience
The intended audience for this course includes:
- Data scientists interested in mining information from images and/or video
- Machine Learning enthusiasts with an interest in computer vision
- Developers interested in learning how to integrate image and video analysis into their own applications
- Anyone interested in learning how Amazon Rekognition works
Learning Objectives
By completing this course, you will:
- Understand what Amazon Rekognition is and what it offers
- Understand the benefits of using the Amazon Rekognition service
- Understand how to use Amazon Rekognition APIs to process both images and videos
- Understand how to use Collections and the storage based API set
- Understand business use cases and scenarios that can benefit from using the Amazon Rekognition service
- Be able to architect and integrate Amazon Rekognition into your own applications
Pre-requisites
The following prerequisites will be both useful and helpful for this course:
- General development and coding experience
- AWS S3 and IAM security experience (for the demonstrations)
Course Agenda
The agenda for the remainder of this course is as follows:
- We’ll discuss what Amazon Rekognition is and when and why you might consider using it
- We’ll review the Amazon Rekognition service and provide an in-depth review of each of its features
- We’ll discuss benefits and business use cases that can be empowered by leveraging Amazon Rekognition
- We’ll provide some example business use cases and scenarios that utilise the Rekognition service
- Finally - We’ll present 2 demonstrations that highlight how to integrate with the Rekognition service
Feedback
If you have thoughts or suggestions for this course, please contact Cloud Academy at support@cloudacademy.com.
Welcome back. In this lecture, we'll cover off a few sample business scenarios where Rekognition can be used. In this first example, a static web app is served up from CloudFront and S3. The web app allows a user to take a picture of themselves using the HTML5 video capability. The web app uses the AWS JavaScript SDK and utilizes Cognito for authentication.
The picture is uploaded into S3 and then Rekognition is used to determine whether the user exists within your collection or not. If a match is found, the user's details are retrieved from a DynamoDB database. The user is allowed into the office, at the same time, the user's development environment is started up using the CloudFormation service. In this example, an IoT camera is used to detect movement.
If movement is detected, an image is taken and uploaded into S3. A call is then made via API gateway to a back-end lambda function. The lambda function invokes Rekognition to perform object and feature extraction to determine what caused the detected motion. Based on results returned from the Rekognition service, the lambda function will write a message to an SNS topic.
An SMS message will then be sent to a subscribing mobile phone to alert the user. The user can then dial back into the service. In this example, we have a mobile lab that allows users to take photos of purchase receipts. The photos are uploaded into S3. This triggers a lambda function which in turn kicks off a state function state machine.
The state machine orchestrates the workflow whereby Rekognition is used to perform text extraction on the purchase receipt. The extracted text is then filtered, formatted and recorded into both DynamoDB and Elastic Search databases. Users can then perform reporting on the monthly expenditure patterns etc. via their mobile app which dials back into the DynamoDB and Elastic Search databases.
That concludes our lecture on some example business scenarios where the Rekognition service could be leveraged. In the next two lectures, we'll provide demonstrations involving using the Rekognition service. The first demonstration involves a static web app that performs facial analysis on photos taken from within the browser. The second demonstration shows a simple use case in which the AWS CLI is used to perform feature and object detection via Rekognition piping the results through various other command line utilities. Go ahead and close this lecture and we'll see you shortly in the next one.
Jeremy is a Content Lead Architect and DevOps SME here at Cloud Academy where he specializes in developing DevOps technical training documentation.
He has a strong background in software engineering, and has been coding with various languages, frameworks, and systems for the past 25+ years. In recent times, Jeremy has been focused on DevOps, Cloud (AWS, Azure, GCP), Security, Kubernetes, and Machine Learning.
Jeremy holds professional certifications for AWS, Azure, GCP, Terraform, Kubernetes (CKA, CKAD, CKS).