Exploring Amazon Rekognition and Textract in .NET Blazor

Introduction

• Amazon Rekognition
• Amazon Textract

Preparation

1. Your favorite IDE.
2. .NET SDK
3. AWS Account (of course, you will need an AWS Account)
4. That's all. :)

Creating the Blazor Project

From Visual Studio 2022

• You may click "Create a new project".

• Search "blazor" and select "Blazor Web App". Click Next.

• Feel free to give any project names and select the project directory. Click Next.

• I use the default value. Feel free to modify the options. Click Create.

For Visual Studio Code

Install the Amazon Recognition and Amazon Textract SDK

For Visual Studio

• Right-click on the Project and click Manage NuGet Packages...

• Click Browse, search Rekognition, click AWSSDK.Rekognition, click Install.

• Search Textract, and click AWSSDK.Textract, click Install.

For Visual Studio Code

• dotnet add RekognitionTextractExploration package AWSSDK.Rekognition
• dotnet add RekognitionTextractExploration package AWSSDK.Textract

Creating a Blazor Page for Amazon Rekognition

• You can right-click on the Pages directory. You may locate the Pages directory under the Components directory. Select Add > New Item. Click on Razor Component and give any appropriate name, for example, DetectLabels. We will learn to detect labels from a picture which is a feature that Amazon Rekognition has.
• You may copy-paste this code to DetectLabels.razor.

• To add the navigation, please add these codes in NavMenu.razor. You may find the file under Components > Layout directory.

• Your NavMenu.razor will be like this.

Explanation

1. The user will upload the file to the server. We will store the file in the temporary directory.
2. We will read the file as a stream and send that to DetectLabelsAsync.
3. After getting the result, we will add the detected labels to the results variable.

Short Demo

Creating a Blazor Page for Amazon Textract

• Create a new Blazor Page for exploring Amazon Textract, for example, Textract.razor.
• You may copy-paste these codes.

• To add the navigation, please add these codes in NavMenu.razor.

Explanation

1. The user will upload the file to the server. We will store the file in the temporary directory.
2. We will read the file as a stream and send that to DetectDocumentTextAsync.
3. After getting the result, we will add the detected texts to the results variable.

Demo

What's your opinion?

Image by katemangostar on Freepik