Create Computer Vision Resource – Computer Vision

Create Computer Vision Resource
In this section, you will learn how to create a computer vision resource using the Azure portal, and after the computer vision service instance is created, you will copy the key and endpoint and save them for later use.

1. Log in to the Azure portal. Select Create Resource. Under the AI+Machine Learning category, select Computer Vision, as shown in Figure 4-1.

Figure 4-1Finding the Computer Vision service

2. Give a subscription, resource group, region, resource name, and pricing tier, then click Create as shown in Figure 4-2.

Figure 4-2Select region, resource name, and pricing tier

3. After a few minutes, you would get your newly deployed Computer Vision service instance as shown in Figure 4-3.

Figure 4-3Newly provisioned computer vision resource

4. Under Resource Management, copy Key 1 and Endpoint as shown in Figure 4-4.

Figure 4-4Copy the Key 1 and Endpoint

Connect a Console App to Computer Vision Resource

1. Download the code provided by Microsoft from the link:
   https://github.com/Azure-Samples/cognitive-services-quickstart-code/blob/master/dotnet/ComputerVision/REST/CSharp-analyze.md#handwriting-recognition-c-example
2. Now, create a new C# console app in Visual Studio as shown in Figure 4-5 and Figure 4-6.

Figure 4-5Select the Console App template in Visual Studio

Figure 4-6Provisioning a console application

3. Choose .NET 6.0 (Long-term support) as the Framework option as shown in Figure 4-7.

Figure 4-7Choose .NET 6.0

4. Using the NuGet Package Manager, add the Newtonsoft.Json package as shown in Figure 4-8.

Figure 4-8Add Newtonsoft.Json using the NuGet Package Manager

5. Under Browse, click Install on Newtonsoft.Json as shown in Figure 4-9.

Figure 4-9Installing the latest stable version of Newtonsoft.Json

6. Replace the Subscription Key and Endpoint from our Azure Portal and paste them into the Program.cs file as shown in Listing 4-1.

namespace CSHttpClientSample{
static class Program {
//Add your Computer Vision subscription key and base endpoint.
static string subscriptionKey = “PASTE_YOUR_COMPUTER_VISION_Resource_KEY_HERE”;
static string endpoint = “PASTE_YOUR_COMPUTER_VISION_ENDPOINT_HERE”;
Listing 4-1Replace the Key and Endpoint, which you copied from your recently deployed Computer Vision service

7. Also, instead of analyze, you must call the OCR endpoint as shown in Listing 4-2.

// the Analyze method endpoint
static string uriBase = endpoint + “vision/v3.1/ocr”;
Listing 4-2Changing endpoint to ocr

8. Get some images for the Vision API to parse. I have my two images here as shown in Figure 4-10. You can download a few images from the given link here:
   https://github.com/Azure-Samples/cognitive-services-sample-data-files/tree/master/ComputerVision/Images

Figure 4-10Images used to be tested by the Vision API

9. Set the “imageFilePath” variable to the image’s location and run the code as shown in Listing 4-3.

//the Analyze method endpoint
static string uriBase = endpoint + “vision/v3.1/ocr”;
// Image you want to be analyzed (add to your bin/debug/netcoreappX.X folder)
static string imageFilePath = @”D:\Data\handwritten.jpg”;
Listing 4-3Providing an image path in the code to be parsed for

10. When you run the application, you will see the output as shown in Listing 4-4 for the image “handwritten.jpg.”

Response:{
“language”:”en”,
“textAngle”:-0.07155849933176689,
“orientation”:”Up”,
“regions”:[
{
“boundingBox”:”96,88,330,373″,
“lines”:[
{
“boundingBox”:”96,88,210,121″,
“words”:
{
“boundingBox”:”96,98,145,111″,
“text”:”LIU”
},
{
“boundingBox”:”246,88,60,97″,
“text”:”T
}
]
},
{
“boundingBox”:”106,239,320,101″,
“words”:[
{
“boundingBox”:”106,247,111,90″,
“text”:”we”
}
]
}
]
}
]
}
Listing 4-4Output for the image “handwritten.jpg”

11. Change the path to the other image, “text.jpg,” as shown in Listing 4-5, and then run the application.

//the Analyze method endpoint
static string uriBase = endpoint + “vision/v3.1/ocr”;
// Image you want to be analyzed (add to your bin/debug/netcoreappX.X folder)
static string imageFilePath = @”D:\Data\text.jpg”;
Listing 4-5Passing another image named “text.jpg” to the code

12. When you execute the program, you will see the output shown below in the Listing 4-6 for the image “text.jpg.”

Response:{
“lanugage”:”en”,
“textAngle”:0.0,
“orientation”:”Up”, “regions”:[
{
“boundingBox”:”20,26,1111,376″,
“lines”:[
{
“boundingBox”:”21,26,184,33″,
“words”:[
{
“boundingBox”:”21,26,184,33″,
“text”:”Microsoft”
},
]
},
{
“boundingBox”:”22,91,326,18″,
“words”:[
{
“boundingBox”:”22,91,43,14″,
“text”:”From”
},
{
“boundingBox”:”71,91,88,18″,
“text”:”Wikipedia”
}
]
}
]
}
]
}
Listing 4-6Output for the image “text.jpg”