Classify Images Machine Learning

Introduction

• Author: Nguyen Trung Nhan
• Based on ML.NET sample at https://github.com/dotnet/machinelearning-samples/blob/master/README.md
• My git: https://github.com/nhannt201

How to use this form?

• You need to download additional data and extract in \bin\Debug\netcoreapp3.1
  Download: https://www.mediafire.com/file/gp8u7a2c0jxeaad/netcoreapp3.1.zip/file
  To create machine learning data to classify images, see more at: https://dotnet.microsoft.com/learn/ml-dotnet/get-started-tutorial/intro

Instructions for using this form

Replace C:\Users\lenovo\Desktop\Scan\ to folder containing the images to your classification!
Replace C:\Users\lenovo\Desktop\Result\ to Result folder in the Guide folder

This sample identification:

• People (ConNguoi)
• Animals (DongVat)
• Items (DoVat)
• Flowers (Hoa)
• Document (TaiLieu)
• Nature (ThienNhien)
• Food (ThucAn)

You must not rename in the Result folder, the classification may be corrupted!
You can Console.WriteLine Name of images with result.Prediction

Example of using this form:

First, the image will be loaded into the program:
  var input = new ModelInput ();
  input.ImageSource = <your_image_url>;
  // Load model and predict output of sample data
  ModelOutput result = ConsumeModel.Predict (input);
Finally, result.Prediction is the name of the image classified based on the above 7 neu characteristics.

Goodluck!