Example Non Maximum Suppression - Revision history

Peter at 23:10, 2 September 2022

2022-09-02T23:10:50Z

Peter at 21:06, 17 January 2022

2022-01-17T21:06:43Z

Peter at 17:51, 12 July 2021

2021-07-12T17:51:21Z

← Older revision		Revision as of 17:51, 12 July 2021
Line 12:		Line 12:
	Non-maximum suppression is a class of algorithm used to find local peaks and minimums inside a feature intensity image. This example demonstrations how to use efficient algorithms inside of BoofCV to quickly find extremes.		Non-maximum suppression is a class of algorithm used to find local peaks and minimums inside a feature intensity image. This example demonstrations how to use efficient algorithms inside of BoofCV to quickly find extremes.
	Example Code:		Example Code:
	* [https://github.com/lessthanoptimal/BoofCV/blob/v0.37/examples/src/main/java/boofcv/examples/features/ExampleNonMaximumSupression.java ExampleNonMaximumSupression.java]		* [https://github.com/lessthanoptimal/BoofCV/blob/v0.38/examples/src/main/java/boofcv/examples/features/ExampleNonMaximumSupression.java ExampleNonMaximumSupression.java]

	Concepts:		Concepts:
Line 25:		Line 25:
	<syntaxhighlight lang="java">		<syntaxhighlight lang="java">
	/**		/**
	* Non-maximum suppression is used to identify local maximums and/or minimums in an image feature intensity map. This		* Non-maximum suppression is used to identify local maximums and/or minimums in an image feature intensity map. This
	* is a common step in feature detection. BoofCV includes an implementation of non-maximum suppression which is much		* is a common step in feature detection. BoofCV includes an implementation of non-maximum suppression which is much
	* faster than the naive algorithm that is often used because of its ease of implementation. The following code		* faster than the naive algorithm that is often used because of its ease of implementation. The following code
	* demonstrates how some of the tuning parameters affects the final output.		* demonstrates how some of the tuning parameters affects the final output.
	*		*
Line 38:		Line 38:
	NonMaxSuppression nonmax = FactoryFeatureExtractor.nonmax(new ConfigExtract(radius, threshold ));		NonMaxSuppression nonmax = FactoryFeatureExtractor.nonmax(new ConfigExtract(radius, threshold ));

	// We will only searching for the maximums. Other variants will look for minimums or will exclude previous		// We will only searching for the maximums. Other variants will look for minimums or will exclude previous
	// candidate detections from being detected twice		// candidate detections from being detected twice
	QueueCorner maximums = new QueueCorner();		QueueCorner maximums = new QueueCorner();

Peter at 20:14, 21 December 2020

2020-12-21T20:14:48Z

← Older revision		Revision as of 20:14, 21 December 2020
Line 12:		Line 12:
	Non-maximum suppression is a class of algorithm used to find local peaks and minimums inside a feature intensity image. This example demonstrations how to use efficient algorithms inside of BoofCV to quickly find extremes.		Non-maximum suppression is a class of algorithm used to find local peaks and minimums inside a feature intensity image. This example demonstrations how to use efficient algorithms inside of BoofCV to quickly find extremes.
	Example Code:		Example Code:
	* [https://github.com/lessthanoptimal/BoofCV/blob/v0.37/examples/src/main/java/boofcv/examples/features/ExampleNonMaximumSupression.java]		* [https://github.com/lessthanoptimal/BoofCV/blob/v0.37/examples/src/main/java/boofcv/examples/features/ExampleNonMaximumSupression.java ExampleNonMaximumSupression.java]

	Concepts:		Concepts:

Peter at 20:14, 21 December 2020

2020-12-21T20:14:31Z

← Older revision		Revision as of 20:14, 21 December 2020
Line 12:		Line 12:
	Non-maximum suppression is a class of algorithm used to find local peaks and minimums inside a feature intensity image. This example demonstrations how to use efficient algorithms inside of BoofCV to quickly find extremes.		Non-maximum suppression is a class of algorithm used to find local peaks and minimums inside a feature intensity image. This example demonstrations how to use efficient algorithms inside of BoofCV to quickly find extremes.
	Example Code:		Example Code:
	* [https://github.com/lessthanoptimal/BoofCV/blob/v0.33/examples/src/main/java/boofcv/examples/features/ExampleNonMaximumSupression.java]		* [https://github.com/lessthanoptimal/BoofCV/blob/v0.37/examples/src/main/java/boofcv/examples/features/ExampleNonMaximumSupression.java]

	Concepts:		Concepts:
Line 58:		Line 58:

	public static void main(String[] args) {		public static void main(String[] args) {

	BufferedImage buffered = UtilImageIO.loadImage(UtilIO.pathExample("standard/boat.jpg"));		BufferedImage buffered = UtilImageIO.loadImage(UtilIO.pathExample("standard/boat.jpg"));

Peter at 13:48, 15 March 2019

2019-03-15T13:48:07Z

← Older revision		Revision as of 13:48, 15 March 2019
Line 12:		Line 12:
	Non-maximum suppression is a class of algorithm used to find local peaks and minimums inside a feature intensity image. This example demonstrations how to use efficient algorithms inside of BoofCV to quickly find extremes.		Non-maximum suppression is a class of algorithm used to find local peaks and minimums inside a feature intensity image. This example demonstrations how to use efficient algorithms inside of BoofCV to quickly find extremes.
	Example Code:		Example Code:
	* [https://github.com/lessthanoptimal/BoofCV/blob/v0.26/examples/src/boofcv/examples/features/~~ExampleNonMaximumSupression.java~~ ExampleNonMaximumSupression.java]		* [https://github.com/lessthanoptimal/BoofCV/blob/v0.33/examples/src/main/java/boofcv/examples/features/ExampleNonMaximumSupression.java]

	Concepts:		Concepts:

Peter at 02:59, 8 December 2016

2016-12-08T02:59:03Z

← Older revision		Revision as of 02:59, 8 December 2016
Line 12:		Line 12:
	Non-maximum suppression is a class of algorithm used to find local peaks and minimums inside a feature intensity image. This example demonstrations how to use efficient algorithms inside of BoofCV to quickly find extremes.		Non-maximum suppression is a class of algorithm used to find local peaks and minimums inside a feature intensity image. This example demonstrations how to use efficient algorithms inside of BoofCV to quickly find extremes.
	Example Code:		Example Code:
	* [https://github.com/lessthanoptimal/BoofCV/blob/v0.26/examples/src/boofcv/examples/~~feature~~/ExampleNonMaximumSupression.java ExampleNonMaximumSupression.java]		* [https://github.com/lessthanoptimal/BoofCV/blob/v0.26/examples/src/boofcv/examples/features/ExampleNonMaximumSupression.java ExampleNonMaximumSupression.java]

	Concepts:		Concepts:

Peter: Created page with " file:Standard_boat.jpg|Input image file:Example_nonmax_intensity...."

2016-12-08T02:57:55Z

Created page with "<center> <gallery caption="Non-maximum suppression with different settings" heights=250 widths=250 perrow=4 > file:Standard_boat.jpg|Input image file:Example_nonmax_intensity...."

New page

<center>
<gallery caption="Non-maximum suppression with different settings" heights=250 widths=250 perrow=4 >
file:Standard_boat.jpg|Input image
file:Example_nonmax_intensity.png|Corner feature intensity
file:Example_nonmax_r3.png|Radius 3
file:Example_nonmax_r3_threshold.png|Radius 3 and Thresholding
file:Example_nonmax_r10.png|Radius 10
file:Example_nonmax_r10_threshold.png|Radius 10 and Thresholding
</gallery>
</center>

Non-maximum suppression is a class of algorithm used to find local peaks and minimums inside a feature intensity image. This example demonstrations how to use efficient algorithms inside of BoofCV to quickly find extremes.
Example Code:
* [https://github.com/lessthanoptimal/BoofCV/blob/v0.26/examples/src/boofcv/examples/feature/ExampleNonMaximumSupression.java ExampleNonMaximumSupression.java]

Concepts:
* Feature detection

Related Examples:
* [[Example_Detect_Interest_Points| Interest Point Detection]]
* [[Tutorial_Image_Segmentation| Tutorial Image Segmentation]]

= Example Code =

<syntaxhighlight lang="java">
/**
* Non-maximum suppression is used to identify local maximums and/or minimums in an image feature intensity map. This
* is a common step in feature detection. BoofCV includes an implementation of non-maximum suppression which is much
* faster than the naive algorithm that is often used because of its ease of implementation. The following code
* demonstrates how some of the tuning parameters affects the final output.
*
* @author Peter Abeles
*/
public class ExampleNonMaximumSupression {

public static BufferedImage renderNonMax( GrayF32 intensity, int radius , float threshold) {
// Create and configure the feature detector
NonMaxSuppression nonmax = FactoryFeatureExtractor.nonmax(new ConfigExtract(radius, threshold ));

// We will only searching for the maximums. Other variants will look for minimums or will exclude previous
// candidate detections from being detected twice
QueueCorner maximums = new QueueCorner();
nonmax.process(intensity, null, null, null, maximums );

// Visualize the intensity image
BufferedImage output = new BufferedImage(intensity.width,intensity.height, BufferedImage.TYPE_INT_RGB);
VisualizeImageData.colorizeSign(intensity, output, -1);

// render each maximum with a circle
Graphics2D g2 = output.createGraphics();
g2.setColor(Color.blue);
for (int i = 0; i < maximums.size(); i++) {
Point2D_I16 c = maximums.get(i);
VisualizeFeatures.drawCircle(g2, c.x, c.y, radius);
}
return output;
}

public static void main(String[] args) {

BufferedImage buffered = UtilImageIO.loadImage(UtilIO.pathExample("standard/boat.jpg"));

GrayF32 input = ConvertBufferedImage.convertFrom(buffered, (GrayF32)null);

// Compute the image gradient
GrayF32 derivX = input.createSameShape();
GrayF32 derivY = input.createSameShape();

GImageDerivativeOps.gradient(DerivativeType.SOBEL, input, derivX, derivY, BorderType.EXTENDED);

// From the gradient compute intensity of shi-tomasi features
GeneralFeatureIntensity<GrayF32,GrayF32> featureIntensity =
FactoryIntensityPoint.shiTomasi(3,false, GrayF32.class);

featureIntensity.process(input, derivX, derivY, null, null , null);
GrayF32 intensity = featureIntensity.getIntensity();

ListDisplayPanel panel = new ListDisplayPanel();
panel.addImage(buffered, "Input Image");
// hack to just show intensity - no features can be detected
panel.addImage(renderNonMax(intensity, 10, Float.MAX_VALUE), "Intensity Image");

// Detect maximums with different settings and visualize the results
panel.addImage(renderNonMax(intensity, 3, -Float.MAX_VALUE), "Radius 3");
panel.addImage(renderNonMax(intensity, 3, 30000), "Radius 3 threshold");
panel.addImage(renderNonMax(intensity, 20, -Float.MAX_VALUE), "Radius 10");
panel.addImage(renderNonMax(intensity, 20, 30000), "Radius 10 threshold");

ShowImages.showWindow(panel, "Non-Maximum Suppression", true);
}
}
</syntaxhighlight>

← Older revision		Revision as of 23:10, 2 September 2022
Line 12:		Line 12:
	Non-maximum suppression is a class of algorithm used to find local peaks and minimums inside a feature intensity image. This example demonstrations how to use efficient algorithms inside of BoofCV to quickly find extremes.		Non-maximum suppression is a class of algorithm used to find local peaks and minimums inside a feature intensity image. This example demonstrations how to use efficient algorithms inside of BoofCV to quickly find extremes.
	Example Code:		Example Code:
	* [https://github.com/lessthanoptimal/BoofCV/blob/v0.40/examples/src/main/java/boofcv/examples/features/ExampleNonMaximumSupression.java ExampleNonMaximumSupression.java]		* [https://github.com/lessthanoptimal/BoofCV/blob/v0.41/examples/src/main/java/boofcv/examples/features/ExampleNonMaximumSupression.java ExampleNonMaximumSupression.java]

	Concepts:		Concepts:
Line 28:		Line 28:
	* is a common step in feature detection. BoofCV includes an implementation of non-maximum suppression which is much		* is a common step in feature detection. BoofCV includes an implementation of non-maximum suppression which is much
	* faster than the naive algorithm that is often used because of its ease of implementation. The following code		* faster than the naive algorithm that is often used because of its ease of implementation. The following code
	* demonstrates how ~~some of the~~ tuning parameters affects the final output.		* demonstrates how tuning parameters affects the final output.
	*		*
	* @author Peter Abeles		* @author Peter Abeles
Line 38:		Line 38:
	NonMaxSuppression nonmax = FactoryFeatureExtractor.nonmax(new ConfigExtract(radius, threshold));		NonMaxSuppression nonmax = FactoryFeatureExtractor.nonmax(new ConfigExtract(radius, threshold));

	// We will only ~~searching~~ for the maximums. Other variants will look for minimums or will exclude previous		// We will only search for the maximums. Other variants will look for minimums or will exclude previous
	// candidate detections from being detected twice		// candidate detections from being detected twice
	~~QueueCorner~~ maximums = new QueueCorner();		var maximums = new QueueCorner();
	nonmax.process(intensity, null, null, null, maximums);		nonmax.process(intensity, null, null, null, maximums);

	// Visualize the intensity image		// Visualize the intensity image
	~~BufferedImage~~ output = new BufferedImage(intensity.width, intensity.height, BufferedImage.TYPE_INT_RGB);		var output = new BufferedImage(intensity.width, intensity.height, BufferedImage.TYPE_INT_RGB);
	VisualizeImageData.colorizeSign(intensity, output, -1);		VisualizeImageData.colorizeSign(intensity, output, -1);

← Older revision		Revision as of 21:06, 17 January 2022
Line 12:		Line 12:
	Non-maximum suppression is a class of algorithm used to find local peaks and minimums inside a feature intensity image. This example demonstrations how to use efficient algorithms inside of BoofCV to quickly find extremes.		Non-maximum suppression is a class of algorithm used to find local peaks and minimums inside a feature intensity image. This example demonstrations how to use efficient algorithms inside of BoofCV to quickly find extremes.
	Example Code:		Example Code:
	* [https://github.com/lessthanoptimal/BoofCV/blob/v0.38/examples/src/main/java/boofcv/examples/features/ExampleNonMaximumSupression.java ExampleNonMaximumSupression.java]		* [https://github.com/lessthanoptimal/BoofCV/blob/v0.40/examples/src/main/java/boofcv/examples/features/ExampleNonMaximumSupression.java ExampleNonMaximumSupression.java]

	Concepts:		Concepts:
Line 34:		Line 34:
	public class ExampleNonMaximumSupression {		public class ExampleNonMaximumSupression {

	public static BufferedImage renderNonMax( GrayF32 intensity, int radius , float threshold) {		public static BufferedImage renderNonMax( GrayF32 intensity, int radius, float threshold ) {
	// Create and configure the feature detector		// Create and configure the feature detector
	NonMaxSuppression nonmax = FactoryFeatureExtractor.nonmax(new ConfigExtract(radius, threshold ));		NonMaxSuppression nonmax = FactoryFeatureExtractor.nonmax(new ConfigExtract(radius, threshold));

	// We will only searching for the maximums. Other variants will look for minimums or will exclude previous		// We will only searching for the maximums. Other variants will look for minimums or will exclude previous
	// candidate detections from being detected twice		// candidate detections from being detected twice
	QueueCorner maximums = new QueueCorner();		QueueCorner maximums = new QueueCorner();
	nonmax.process(intensity, null, null, null, maximums );		nonmax.process(intensity, null, null, null, maximums);

	// Visualize the intensity image		// Visualize the intensity image
	BufferedImage output = new BufferedImage(intensity.width,intensity.height, BufferedImage.TYPE_INT_RGB);		BufferedImage output = new BufferedImage(intensity.width, intensity.height, BufferedImage.TYPE_INT_RGB);
	VisualizeImageData.colorizeSign(intensity, output, -1);		VisualizeImageData.colorizeSign(intensity, output, -1);

Line 57:		Line 57:
	}		}

	public static void main(String[] args) {		public static void main( String[] args ) {
	BufferedImage buffered = UtilImageIO.~~loadImage~~(UtilIO.pathExample("standard/boat.jpg"));		BufferedImage buffered = UtilImageIO.loadImageNotNull(UtilIO.pathExample("standard/boat.jpg"));

	GrayF32 input = ConvertBufferedImage.convertFrom(buffered, (GrayF32)null);		GrayF32 input = ConvertBufferedImage.convertFrom(buffered, (GrayF32)null);
Line 69:		Line 69:

	// From the gradient compute intensity of shi-tomasi features		// From the gradient compute intensity of shi-tomasi features
	GeneralFeatureIntensity<GrayF32,GrayF32> featureIntensity =		GeneralFeatureIntensity<GrayF32, GrayF32> featureIntensity =
	FactoryIntensityPoint.shiTomasi(3,false, GrayF32.class);		FactoryIntensityPoint.shiTomasi(3, false, GrayF32.class);

	featureIntensity.process(input, derivX, derivY, null, null , null);		featureIntensity.process(input, derivX, derivY, null, null, null);
	GrayF32 intensity = featureIntensity.getIntensity();		GrayF32 intensity = featureIntensity.getIntensity();

Line 78:		Line 78:
	panel.addImage(buffered, "Input Image");		panel.addImage(buffered, "Input Image");
	// hack to just show intensity - no features can be detected		// hack to just show intensity - no features can be detected
	panel.addImage(renderNonMax(intensity, 10, Float.MAX_VALUE), "Intensity Image");		panel.addImage(renderNonMax(intensity, 10, Float.MAX_VALUE), "Intensity Image");

	// Detect maximums with different settings and visualize the results		// Detect maximums with different settings and visualize the results
	panel.addImage(renderNonMax(intensity, 3, -Float.MAX_VALUE), "Radius 3");		panel.addImage(renderNonMax(intensity, 3, -Float.MAX_VALUE), "Radius 3");
	panel.addImage(renderNonMax(intensity, 3, 30000), "Radius 3 threshold");		panel.addImage(renderNonMax(intensity, 3, 30000), "Radius 3 threshold");
	panel.addImage(renderNonMax(intensity, 20, -Float.MAX_VALUE), "Radius 10");		panel.addImage(renderNonMax(intensity, 20, -Float.MAX_VALUE), "Radius 10");
	panel.addImage(renderNonMax(intensity, 20, 30000), "Radius 10 threshold");		panel.addImage(renderNonMax(intensity, 20, 30000), "Radius 10 threshold");

	ShowImages.showWindow(panel, "Non-Maximum Suppression", true);		ShowImages.showWindow(panel, "Non-Maximum Suppression", true);