GraphCutMaskingOptions

GraphCutMaskingOptions class

The GraphCut auto masking options.

public class GraphCutMaskingOptions : MaskingOptions

Constructors

Name Description
GraphCutMaskingOptions() The default constructor.

Properties

Name Description
Args { get; set; } Gets or sets the arguments for segmentation algorithm.
BackgroundReplacementColor { get; set; } Gets or sets the background replacement color.
Decompose { get; set; } Gets or sets a value indicating whether needless to separate each Shape from mask as individual object or as united object from mask separated from background.
ExportOptions { get; set; } Gets or sets the image export options.
FeatheringRadius { get; set; } Gets or sets the feathering radius.
MaskingArea { get; set; } Gets or sets the masking area.
Method { get; set; } Gets or sets the segmentation method.

Examples

Saving image masking result with feathering based on image size. Image masking is performed using auto calculated default strokes. The Args property of AutoMaskingGraphCutOptions can be omitted since default strokes are placed there in the end.

[C#]

MaskingResult[] results;
using (RasterImage image = (RasterImage)Image.Load("input.jpg"))
{
    AutoMaskingGraphCutOptions options = new AutoMaskingGraphCutOptions
                                                {
                                                    CalculateDefaultStrokes = true,
                                                    FeatheringRadius = (Math.Max(image.Width, image.Height) / 500) + 1,
                                                    Method = SegmentationMethod.GraphCut,
                                                    Decompose = false,
                                                    ExportOptions =
                                                        new PngOptions()
                                                            {
                                                                ColorType = PngColorType.TruecolorWithAlpha,
                                                                Source = new FileCreateSource("tempFile")
                                                            },
                                                    BackgroundReplacementColor = Color.Transparent
    };

    results = new ImageMasking(image).Decompose(options);
}

using (RasterImage resultImage = (RasterImage)results[1].GetImage())
{
    resultImage.Save("output.png", new PngOptions() { ColorType = PngColorType.TruecolorWithAlpha });
}

Saving image masking result with feathering based on image size. Image masking is performed using auto calculated default strokes. Additionally the data of the two assumed objects is also specified in the AssumedObjects property of the AutoMaskingGraphCutOptions.

[C#]

List<AssumedObjectData> assumedObjects = new List<AssumedObjectData>();
assumedObjects.Add(new AssumedObjectData(DetectedObjectType.Human, new Rectangle(100, 100, 150, 300)));
assumedObjects.Add(new AssumedObjectData(DetectedObjectType.Dog, new Rectangle(300, 100, 50, 30)));

MaskingResult[] results;
using (RasterImage image = (RasterImage)Image.Load("input.jpg"))
{
    AutoMaskingGraphCutOptions options = new AutoMaskingGraphCutOptions
                                                {
                                                    AssumedObjects = assumedObjects,
                                                    CalculateDefaultStrokes = true,
                                                    FeatheringRadius = (Math.Max(image.Width, image.Height) / 500) + 1,
                                                    Method = SegmentationMethod.GraphCut,
                                                    Decompose = false,
                                                    ExportOptions =
                                                        new PngOptions()
                                                            {
                                                                ColorType = PngColorType.TruecolorWithAlpha,
                                                                Source = new FileCreateSource("tempFile")
                                                            },
                                                    BackgroundReplacementColor = Color.Transparent
    };

    results = new ImageMasking(image).Decompose(options);
}

using (RasterImage resultImage = (RasterImage)results[1].GetImage())
{
    resultImage.Save("output.png", new PngOptions() { ColorType = PngColorType.TruecolorWithAlpha });
}

Saving Graph Cut image masking result with feathering set to 3. Image masking is performed using specified Point array.

[C#]

MaskingResult[] results;
using (RasterImage image = (RasterImage)Image.Load("input.jpg"))
{
    GraphCutMaskingOptions options = new GraphCutMaskingOptions()
                                                {
                                                    FeatheringRadius = 3,
                                                    Method = SegmentationMethod.GraphCut,
                                                    Decompose = false,
                                                    ExportOptions =
                                                        new PngOptions()
                                                            {
                                                                ColorType = PngColorType.TruecolorWithAlpha,
                                                                Source = new FileCreateSource("tempFile")
                                                            },
                                                    BackgroundReplacementColor = Color.Transparent,
                                                    Args = new AutoMaskingArgs()
                                                            {
                                                                ObjectsPoints = new Point[][]
                                                                                    {
                                                                                        new Point[]
                                                                                            {
                                                                                                new Point(100, 100),
                                                                                            },
                                                                                    }
                                                            }
                                                };

    results = new ImageMasking(image).Decompose(options);
}

using (RasterImage resultImage = (RasterImage)results[1].GetImage())
{
    resultImage.Save("output.png", new PngOptions() { ColorType = PngColorType.TruecolorWithAlpha });
}

Saving image masking result with feathering based on image size and re-using masking options for the new masking iteration. Image masking is performed using auto calculated default strokes. Additionally the data of the two assumed objects is also specified in the AssumedObjects property of the AutoMaskingGraphCutOptions. After getting initial masking result, applied background/foreground strokes are modified and another masking iteration is performed.

[C#]

List<AssumedObjectData> assumedObjects = new List<AssumedObjectData>();
assumedObjects.Add(new AssumedObjectData(DetectedObjectType.Human, new Rectangle(100, 100, 150, 300)));
assumedObjects.Add(new AssumedObjectData(DetectedObjectType.Dog, new Rectangle(300, 100, 50, 30)));

MaskingResult[] results;
AutoMaskingGraphCutOptions options;
using (RasterImage image = (RasterImage)Image.Load("input.jpg"))
{
    options = new AutoMaskingGraphCutOptions
                    {
                        AssumedObjects = assumedObjects,
                        CalculateDefaultStrokes = true,
                        FeatheringRadius = 3,
                        Method = SegmentationMethod.GraphCut,
                        Decompose = false,
                        ExportOptions =
                            new PngOptions()
                                {
                                    ColorType = PngColorType.TruecolorWithAlpha,
                                    Source = new FileCreateSource("tempFile")
                                },
                        BackgroundReplacementColor = Color.Transparent
                    };

    results = new ImageMasking(image).Decompose(options);
}

// At this point applied foreground/background strokes can be analyzed and based on it additional 
// foreground/background strokes can be manually provided.
Point[] appliedBackgroundStrokes = options.DefaultBackgroundStrokes;
Point[] appliedForegroundStrokes = options.DefaultForegroundStrokes;
Rectangle[] appliedObjectRectangles = options.DefaultObjectsRectangles;
using (RasterImage resultImage = (RasterImage)results[1].GetImage())
{
    resultImage.Save("output.png", new PngOptions() { ColorType = PngColorType.TruecolorWithAlpha });
}

using (RasterImage image = (RasterImage)Image.Load("input.jpg"))
{
    // Re-using AutoMaskingGraphCutOptions there is no need to perform default strokes calculations second time.
    options.CalculateDefaultStrokes = false;
    // When both default strokes and ObjectsPoints in the Args property of AutoMaskingArgs are provided, Point arrays are end up combined.
    // The first ObjectsPoints array is considered to be a background points array and 
    // the second ObjectsPoints array is considered to be a foreground points array.
    // When both DefaultObjectsRectangles and ObjectsRectangles in the Args property of AutoMaskingArgs are provided, 
    // only the array from the Args is being used.
    options.Args = new AutoMaskingArgs()
                        {
                            ObjectsPoints = new Point[][]
                                                {
                                                    new Point[] { new Point(100, 100), new Point(150, 100) }, 
                                                    new Point[] { new Point(500, 200) }, 
                                                },
                            ObjectsRectangles = new Rectangle[]
                                                    {
                                                        new Rectangle(100, 100, 300, 300), 
                                                    }
                        };
    results = new ImageMasking(image).Decompose(options);
}

using (RasterImage resultImage = (RasterImage)results[1].GetImage())
{
    resultImage.Save("output.png", new PngOptions() { ColorType = PngColorType.TruecolorWithAlpha });
}

Saving image masking result with feathering based on image size, modifying obtained default strokes and using it for the new masking iteration. Image masking is performed using auto calculated default strokes. Additionally the data of the two assumed objects is also specified in the AssumedObjects property of the AutoMaskingGraphCutOptions. After getting initial masking result, applied background/foreground strokes are modified and another masking iteration is performed using new GraphCutMaskingOptions instance.

[C#]

List<AssumedObjectData> assumedObjects = new List<AssumedObjectData>();
assumedObjects.Add(new AssumedObjectData(DetectedObjectType.Human, new Rectangle(100, 100, 150, 300)));
assumedObjects.Add(new AssumedObjectData(DetectedObjectType.Dog, new Rectangle(300, 100, 50, 30)));

MaskingResult[] results;
AutoMaskingGraphCutOptions options;

using (RasterImage image = (RasterImage)Image.Load("input.jpg"))
{
    options = new AutoMaskingGraphCutOptions
                    {
                        AssumedObjects = assumedObjects,
                        CalculateDefaultStrokes = true,
                        FeatheringRadius = 3,
                        Method = SegmentationMethod.GraphCut,
                        Decompose = false,
                        ExportOptions =
                            new PngOptions()
                                {
                                    ColorType = PngColorType.TruecolorWithAlpha,
                                    Source = new FileCreateSource("tempFile")
                                },
                        BackgroundReplacementColor = Color.Transparent
                    };

    results = new ImageMasking(image).Decompose(options);
}

// At this point applied foreground/background strokes can be analyzed and based on it additional 
// foreground/background strokes can be manually provided.
Point[] appliedBackgroundStrokes = options.DefaultBackgroundStrokes;
Point[] appliedForegroundStrokes = options.DefaultForegroundStrokes;
Rectangle[] appliedObjectRectangles = options.DefaultObjectsRectangles;
using (RasterImage resultImage = (RasterImage)results[1].GetImage())
{
    resultImage.Save("output.png", new PngOptions() { ColorType = PngColorType.TruecolorWithAlpha });
}

appliedBackgroundStrokes[5] = new Point(100, 100);
appliedBackgroundStrokes[15] = new Point(150, 100);

appliedForegroundStrokes[1] = new Point(500, 200);

appliedObjectRectangles[0] = new Rectangle(100, 100, 300, 300);

using (RasterImage image = (RasterImage)Image.Load("input.jpg"))
{
    GraphCutMaskingOptions graphCutOptions = new GraphCutMaskingOptions()
                                                    {
                                                        FeatheringRadius = 3,
                                                        Method = SegmentationMethod.GraphCut,
                                                        Decompose = false,
                                                        ExportOptions = new PngOptions()
                                                                            {
                                                                                ColorType = PngColorType.TruecolorWithAlpha,
                                                                                Source = new FileCreateSource("tempFile")
                                                                            },
                                                        BackgroundReplacementColor = Color.Transparent,
                                                        Args = new AutoMaskingArgs()
                                                                {
                                                                    ObjectsPoints = new Point[][]
                                                                                        {
                                                                                            appliedBackgroundStrokes,
                                                                                            appliedForegroundStrokes
                                                                                        },
                                                                    ObjectsRectangles = appliedObjectRectangles
                                                                }
                                                    };
    results = new ImageMasking(image).Decompose(graphCutOptions);
}

using (RasterImage resultImage = (RasterImage)results[1].GetImage())
{
    resultImage.Save("output.png", new PngOptions() { ColorType = PngColorType.TruecolorWithAlpha });
}

See Also