Unlocking the Potential of Azure Custom Vision Service: A Step-by-Step Guide

Introduction

Today we will explore how to create a Custom Vision Service in Azure and the different capabilities of Azure Custom Vision Service models like Classification and Object Detection. In this article, we will build and deploy the Azure Custom Vision Service using Classification models.

What is Custom Vision Service?

Custom Vision is a cloud-based machine learning service that Microsoft Azure provides that enables developers to train and deploy custom image classification and object detection models. With Custom Vision, developers can easily upload and label their images, train and fine-tune a model using simple interfaces, and then export it for use in their applications. This service is particularly useful for developers who do not have expertise in machine learning or deep learning but still need to create custom models for their specific use cases.

We have the following steps to create a Custom Vision Service in the Azure Portal.

Go to the Azure portal and sign in with your Azure account.
In the left-hand menu, click on "Create a resource".
Search for "Custom Vision" in the search bar and select "Custom Vision" from the search results.
Click on "Create" on the Custom Vision page.
In the "Create" blade, enter the following details.
Subscription: Select your subscription.
Resource group: Choose a resource group or create a new one.
Name: Enter a name for your Custom Vision service.
Training location: Select the region where you want to train your model.
Prediction location: Select the region where you want to make predictions.
Pricing tier: Choose a pricing tier that suits your needs.
Training data source: Choose either "Azure Blob Storage" or "Azure Files".
After entering the required details, click on "Create" to create your Custom Vision service.
Once your service is created, navigate to it from the Azure portal dashboard.

Let us Start.

Step 1. Login into Azure Portal.

Azure Login Image

Step 2. Search Custom Vision in the search bar.

Azure sevices

Step 3. Click Create Custom Vision.

Custom vision

Step 4. In the Create Custom Vision page, enter the details as mentioned below.

Create Custom vision

Step 5. Click the Review + Create button, and you will get a validation passed message as a status.

Validation Passed

Step 6. Click Create button. Deployment started initializing in a minute or two this became successful.

Cognitive Services

Step 7. Goto Resource Group.

Resource Group

General Steps to Create Custom Vision Solution in Azure

Create a new project: In the Custom Vision portal, create a new project and select the type of project you want to create (classification or object detection).
Add and label images: Add images to the project and label them with the appropriate tags to indicate the object or classification.
Train the model: Train the model by selecting the appropriate training parameters and initiating the training process.
Test and improve the model: Test the model with additional images and adjust the training parameters as needed to improve the model's accuracy.

Step 1. Login into the Custom Vision portal and then click the Sign in button.

customvision.ai

Step 2. Custom Vision portal appeared on the screen as follows, and then click Create a new project.

New Project

Step 3. Add the images and name the images with proper tagging into a category.

Add image

Images are uploaded and tagged with proper names.

Step 4. Train the model and choose the training type as Quick Training.

Choose traning type

Project1

Step 5. Test the model either you can upload the images locally or browse the images from the URL.

Quick Test

The above image was successfully tested with 100% Probability.

Conclusion

I hope you enjoyed reading the article. This article taught us about creating and building Custom Vision projects using Microsoft Azure. Overall, Custom Vision Azure projects can allow organizations to automate image recognition tasks and gain valuable insights from visual data, leading to improved efficiency, productivity, and accuracy.