ImageEstimatorsCatalog.ConvertToImage Method

Definition

Create a VectorToImageConvertingEstimator, which creates image from the data from the column specified in inputColumnName to a new column: outputColumnName.

public static Microsoft.ML.Transforms.Image.VectorToImageConvertingEstimator ConvertToImage (this Microsoft.ML.TransformsCatalog catalog, int imageHeight, int imageWidth, string outputColumnName, string inputColumnName = default, Microsoft.ML.Transforms.Image.ImagePixelExtractingEstimator.ColorBits colorsPresent = Microsoft.ML.Transforms.Image.ImagePixelExtractingEstimator+ColorBits.Rgb, Microsoft.ML.Transforms.Image.ImagePixelExtractingEstimator.ColorsOrder orderOfColors = Microsoft.ML.Transforms.Image.ImagePixelExtractingEstimator+ColorsOrder.ARGB, bool interleavedColors = false, float scaleImage = 1, float offsetImage = 0, int defaultAlpha = 255, int defaultRed = 0, int defaultGreen = 0, int defaultBlue = 0);
static member ConvertToImage : Microsoft.ML.TransformsCatalog * int * int * string * string * Microsoft.ML.Transforms.Image.ImagePixelExtractingEstimator.ColorBits * Microsoft.ML.Transforms.Image.ImagePixelExtractingEstimator.ColorsOrder * bool * single * single * int * int * int * int -> Microsoft.ML.Transforms.Image.VectorToImageConvertingEstimator
<Extension()>
Public Function ConvertToImage (catalog As TransformsCatalog, imageHeight As Integer, imageWidth As Integer, outputColumnName As String, Optional inputColumnName As String = Nothing, Optional colorsPresent As ImagePixelExtractingEstimator.ColorBits = Microsoft.ML.Transforms.Image.ImagePixelExtractingEstimator+ColorBits.Rgb, Optional orderOfColors As ImagePixelExtractingEstimator.ColorsOrder = Microsoft.ML.Transforms.Image.ImagePixelExtractingEstimator+ColorsOrder.ARGB, Optional interleavedColors As Boolean = false, Optional scaleImage As Single = 1, Optional offsetImage As Single = 0, Optional defaultAlpha As Integer = 255, Optional defaultRed As Integer = 0, Optional defaultGreen As Integer = 0, Optional defaultBlue As Integer = 0) As VectorToImageConvertingEstimator

Parameters

catalog
TransformsCatalog

The transform's catalog.

imageHeight
Int32

The height of the output images.

imageWidth
Int32

The width of the output images.

outputColumnName
String

Name of the column resulting from the transformation of inputColumnName. This column's data type will be MLImage.

inputColumnName
String

Name of the column with data to be converted to image. This estimator operates over known-sized vector of Single, Double and Byte.

colorsPresent
ImagePixelExtractingEstimator.ColorBits

Specifies which ImagePixelExtractingEstimator.ColorBits are in present the input pixel vectors. The order of colors is specified in orderOfColors.

orderOfColors
ImagePixelExtractingEstimator.ColorsOrder

The order in which colors are presented in the input vector.

interleavedColors
Boolean

Whether the pixels are interleaved, meaning whether they are in orderOfColors order, or separated in the planar form: all the values for one color for all pixels, then all the values for another color and so on.

scaleImage
Single

The values are scaled by this value before being converted to pixels. Applied to vector value before offsetImage.

offsetImage
Single

The offset is subtracted before converting the values to pixels. Applied to vector value after scaleImage.

defaultAlpha
Int32

Default value for alpha color, would be overridden if colorsPresent contains Alpha.

defaultRed
Int32

Default value for red color, would be overridden if colorsPresent contains Red.

defaultGreen
Int32

Default value for green color, would be overridden if colorsPresent contains Green.

defaultBlue
Int32

Default value for blue color, would be overridden if colorsPresent contains Blue.

Returns

Examples

using System;
using System.Collections.Generic;
using System.Linq;
using Microsoft.ML;
using Microsoft.ML.Data;

namespace Samples.Dynamic
{
    public static class ConvertToImage
    {
        private const int imageHeight = 224;
        private const int imageWidth = 224;
        private const int numberOfChannels = 3;
        private const int inputSize = imageHeight * imageWidth * numberOfChannels;

        // Sample that shows how an input array (of doubles) can be used to interop
        // with image related estimators in ML.NET.
        public static void Example()
        {
            // Create a new ML context, for ML.NET operations. It can be used for
            // exception tracking and logging, as well as the source of randomness.
            var mlContext = new MLContext();

            // Create a list of training data points.
            var dataPoints = GenerateRandomDataPoints(4);

            // Convert the list of data points to an IDataView object, which is
            // consumable by ML.NET API.
            var data = mlContext.Data.LoadFromEnumerable(dataPoints);

            // Image loading pipeline.
            var pipeline = mlContext.Transforms.ConvertToImage(imageHeight,
                imageWidth, "Image", "Features")
                .Append(mlContext.Transforms.ExtractPixels("Pixels", "Image"));

            var transformedData = pipeline.Fit(data).Transform(data);

            // Preview the transformedData.
            PrintColumns(transformedData);

            // Features                 Image                    Pixels
            // 185,209,196,142,52...    {Width=224, Height=224}  185,209,196,142,52...
            // 182,235,84,23,87...      {Width=224, Height=224}  182,235,84,23,87...
            // 192,214,247,22,38...     {Width=224, Height=224}  192,214,247,22,38...
            // 242,161,141,223,192...   {Width=224, Height=224}  242,161,141,223,192...
        }

        private static void PrintColumns(IDataView transformedData)
        {
            Console.WriteLine("{0, -25} {1, -25} {2, -25}", "Features", "Image",
                "Pixels");

            using (var cursor = transformedData.GetRowCursor(transformedData
                .Schema))
            {
                // Note that it is best to get the getters and values *before*
                // iteration, so as to facilitate buffer sharing (if applicable), and
                // column -type validation once, rather than many times.
                VBuffer<float> features = default;
                VBuffer<float> pixels = default;
                MLImage imageObject = null;

                var featuresGetter = cursor.GetGetter<VBuffer<float>>(cursor.Schema[
                    "Features"]);

                var pixelsGetter = cursor.GetGetter<VBuffer<float>>(cursor.Schema[
                    "Pixels"]);

                var imageGetter = cursor.GetGetter<MLImage>(cursor.Schema["Image"]);
                while (cursor.MoveNext())
                {

                    featuresGetter(ref features);
                    pixelsGetter(ref pixels);
                    imageGetter(ref imageObject);

                    Console.WriteLine("{0, -25} {1, -25} {2, -25}", string.Join(",",
                        features.DenseValues().Take(5)) + "...",
                        $"Width={imageObject.Width}, Height={imageObject.Height}",
                        string.Join(",", pixels.DenseValues().Take(5)) + "...");
                }

                // Dispose the image.
                imageObject.Dispose();
            }
        }

        private class DataPoint
        {
            [VectorType(inputSize)]
            public float[] Features { get; set; }
        }

        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
            int seed = 0)
        {
            var random = new Random(seed);

            for (int i = 0; i < count; i++)
                yield return new DataPoint
                {
                    Features = Enumerable.Repeat(0,
                    inputSize).Select(x => (float)random.Next(0, 256)).ToArray()
                };
        }
    }
}

Applies to