Computer Vision API provides ready algorithms for processing images – It can classify a picture as object, animal etc. It can also validate it with Celebrities – all through the machine learning infrastructure of Azure under the hood.
Azure Service
For testing this you need to create the Computer Vision service from Azure > Marketplace.
Choose the S1 tier as most of the capabilities are residing there.
Console Application in C#
Create a Console Application in C# and add the following Nuget package.
SRK
As a test I am going to upload above SRK picture & let the Computer Vision API identify him.
Output
The API is successfully identifying SRK! 
The Code
using Microsoft.Azure.CognitiveServices.Vision.ComputerVision;
using Microsoft.Azure.CognitiveServices.Vision.ComputerVision.Models;
using Microsoft.Rest;
using System;
using System.IO;
using System.Net.Http;
namespace ComputerVisionAPI
{
class Program
{
static void Main(string[] args)
{
var features = new VisualFeatureTypes[] { VisualFeatureTypes.Tags, VisualFeatureTypes.Description };
ComputerVisionClient computerVisionClient = new ComputerVisionClient(
new ApiKeyServiceClientCredentials(“YOUR KEY”),
new DelegatingHandler[] { }
)
{
Endpoint = https://URL.cognitiveservices.azure.com/
};
using (var fs = new FileStream(@”C:\temp\ShahrukhKhan.png”, FileMode.Open))
{
ImageAnalysis result = computerVisionClient.AnalyzeImageInStreamAsync(fs, features).Result;
Console.WriteLine(“TAGS >> “);
foreach (string tag in result.Description.Tags)
Console.Write($” {tag} “);
Console.WriteLine(Environment.NewLine + Environment.NewLine + “CAPTION >>”);
foreach (ImageCaption caption in result.Description.Captions)
Console.WriteLine($”{caption.Text} – Confidence: {caption.Confidence} “);
}
}
}
}
References
https://azure.microsoft.com/en-us/services/cognitive-services/computer-vision/#product-overview
