Package ai.djl.modality.cv.output

Class CategoryMask

java.lang.Object

ai.djl.modality.cv.output.CategoryMask

All Implemented Interfaces:

BytesSupplier, ai.djl.util.JsonSerializable, Serializable

public class CategoryMask
extends Object
implements ai.djl.util.JsonSerializable

A class representing the segmentation result of an image in an Application.CV.SEMANTIC_SEGMENTATION case.

See Also:

• Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static final class	CategoryMask.SegmentationSerializer	A customized Gson serializer to serialize the Segmentation object.

Constructor Summary

Constructors

Constructor	Description
CategoryMask(List<String> classes, int[][] mask)	Constructs a Mask with the given data.

Method Summary

Modifier and Type	Method	Description
void	drawMask(Image image, int opacity)	Highlights the detected object on the image with random colors.
void	drawMask(Image image, int opacity, int background)	Highlights the detected object on the image with random colors.
void	drawMask(Image image, int classId, int color, int opacity)	Highlights the specified object with specific color.
Image	getBackgroundImage(Image image)	Extracts the background from the image.
List<String>	getClasses()	Returns the list of classes.
int[][]	getMask()	Returns the class for each pixel.
Image	getMaskImage(Image image)	Extracts the detected objects from the image.
Image	getMaskImage(Image image, int classId)	Extracts the specified object from the image.
ByteBuffer	toByteBuffer()	Returns the ByteBuffer presentation of the object.
String	toJson()

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface ai.djl.ndarray.BytesSupplier

getAsBytes, getAsObject, getAsString

Constructor Details

CategoryMask

public CategoryMask(List<String> classes,
 int[][] mask)

Constructs a Mask with the given data.

Parameters:

classes - the list of classes

mask - the category mask for each pixel in the image

Method Details

getClasses

public List<String> getClasses()

Returns the list of classes.

Returns:

list of classes

getMask

public int[][] getMask()

Returns the class for each pixel.

Returns:

the class for each pixel

toByteBuffer

public ByteBuffer toByteBuffer()

Returns the ByteBuffer presentation of the object.

Specified by:

toByteBuffer in interface BytesSupplier

Returns:

the ByteBuffer presentation of the object

toJson

public String toJson()

Specified by:

toJson in interface ai.djl.util.JsonSerializable

getMaskImage

public Image getMaskImage(Image image)

Extracts the detected objects from the image.

Parameters:

image - the original image

Returns:

the detected objects from the image

getMaskImage

public Image getMaskImage(Image image,
 int classId)

Extracts the specified object from the image.

Parameters:

image - the original image

classId - the class to extract from the image

Returns:

the specific object on the image

getBackgroundImage

public Image getBackgroundImage(Image image)

Extracts the background from the image.

Parameters:

image - the original image

Returns:

the background of the image

drawMask

public void drawMask(Image image,
 int opacity)

Highlights the detected object on the image with random colors.

Parameters:

image - the original image

opacity - the opacity of the overlay. Value is between 0 and 255 inclusive, where 0 means the overlay is completely transparent and 255 means the overlay is completely opaque.

drawMask

public void drawMask(Image image,
 int opacity,
 int background)

Highlights the detected object on the image with random colors.

Parameters:

image - the original image

opacity - the opacity of the overlay. Value is between 0 and 255 inclusive, where 0 means the overlay is completely transparent and 255 means the overlay is completely opaque.

background - replace the background with specified background color, use transparent color to remove background

drawMask

public void drawMask(Image image,
 int classId,
 int color,
 int opacity)

Highlights the specified object with specific color.

Parameters:

image - the original image

classId - the class to draw on the image

color - the rgb color with opacity

opacity - the opacity of the overlay. Value is between 0 and 255 inclusive, where 0 means the overlay is completely transparent and 255 means the overlay is completely opaque.