Difference between revisions of "Example Binary Image"

From BoofCV
Jump to navigationJump to search
(Updated for v0.3)
m
 
(13 intermediate revisions by the same user not shown)
Line 1: Line 1:
= Binary Image Processing =
<center>
<center>
<gallery caption="Example input and output Images" heights=150 widths=200 >
<gallery caption="Example input and output Images" heights=150 widths=200 >
Line 6: Line 4:
Image:example_binary.png|Thresholded Binary Image.
Image:example_binary.png|Thresholded Binary Image.
Image:example_binary_labeled.png|Labeled Binary Image
Image:example_binary_labeled.png|Labeled Binary Image
Image:example_binary_contour.png|Contour Image
</gallery>
</gallery>
</center>
</center>


Binary images are images where each pixel can take on two values, typically represented by 0 or 1.  Binary images are easy to compute and fast to process, which makes them popular in many applications.
In a binary image each pixel can have a value of 0 or 1.  Binary images are easy to compute and fast to process, which makes them popular in many applications.  BoofCV contains many operations for creating and manipulating binary images.  The example below demonstrates a few of ones contained inside of BinaryImageOps.


Example File: [https://github.com/lessthanoptimal/BoofCV/blob/master/examples/src/boofcv/examples/ExampleBinaryImage.java ExampleBinaryImage.java]
Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.40/examples/src/main/java/boofcv/examples/imageprocessing/ExampleBinaryOps.java ExampleBinaryOps.java]


Concepts:
Concepts:
Line 19: Line 18:
* Pixel Math
* Pixel Math
* Image Rendering
* Image Rendering
Relevant Applets:
* [[Applet_Binary_Operations| Binary Operations]]
* [[Applet_Binary_Segmentation| Binary Segmentation]]


= Basic Example =
= Basic Example =
In this example a threshold is computed for the input image dynamically and the resulting binary image shown.
<syntaxhighlight lang="java">
<syntaxhighlight lang="java">
public static void binaryExample( BufferedImage image )
/**
{
* Demonstrates how to create binary images by thresholding, applying binary morphological operations, and
// convert into a usable format
* then extracting detected features by finding their contours.
ImageFloat32 input = ConvertBufferedImage.convertFromSingle(image, null, ImageFloat32.class);
*
ImageUInt8 binary = new ImageUInt8(input.width,input.height);
* @author Peter Abeles
 
* @see boofcv.examples.segmentation.ExampleThresholding
// the mean pixel value is often a reasonable threshold when creating a binary image
*/
float mean = PixelMath.sum(input)/(input.width*input.height);
public class ExampleBinaryOps {
 
public static void main( String[] args ) {
// create a binary image
// load and convert the image into a usable format
ThresholdImageOps.threshold(input,binary,mean,true);
BufferedImage image = UtilImageIO.loadImageNotNull(UtilIO.pathExample("particles01.jpg"));
 
// Render the binary image for output and display it in a window
BufferedImage visualBinary = VisualizeBinaryData.renderBinary(binary,null);
ShowImages.showWindow(visualBinary,"Binary Image");
}
</syntaxhighlight>
 
= Labeled Example =
Here clustered of blobs are detected and arbitrarily assigned labels.  Noise is reduced through morphological image operations.


<syntaxhighlight lang="java">
public static void labeledExample( BufferedImage image )
{
// convert into a usable format
// convert into a usable format
ImageFloat32 input = ConvertBufferedImage.convertFromSingle(image, null, ImageFloat32.class);
GrayF32 input = ConvertBufferedImage.convertFromSingle(image, null, GrayF32.class);
ImageUInt8 binary = new ImageUInt8(input.width,input.height);
var binary = new GrayU8(input.width, input.height);
ImageSInt32 blobs = new ImageSInt32(input.width,input.height);
var label = new GrayS32(input.width, input.height);


// the mean pixel value is often a reasonable threshold when creating a binary image
// Select a global threshold using Otsu's method.
float mean = PixelMath.sum(input)/(input.width*input.height);
double threshold = GThresholdImageOps.computeOtsu(input, 0, 255);


// create a binary image
// Apply the threshold to create a binary image
ThresholdImageOps.threshold(input,binary,mean,true);
ThresholdImageOps.threshold(input, binary, (float)threshold, true);


// remove small blobs through erosion and dilation
// remove small blobs through erosion and dilation
// The null in the input indicates that it should internally declare the work image it needs
// The null in the input indicates that it should internally declare the work image it needs
// this is less efficient, but easier to code.
// this is less efficient, but easier to code.
binary = BinaryImageOps.erode8(binary,null);
GrayU8 filtered = BinaryImageOps.erode8(binary, 1, null);
binary = BinaryImageOps.dilate8(binary, null);
filtered = BinaryImageOps.dilate8(filtered, 1, null);
 
// Detect blobs inside the image using an 8-connect rule
List<Contour> contours = BinaryImageOps.contour(filtered, ConnectRule.EIGHT, label);
 
// colors of contours
int colorExternal = 0xFFFFFF;
int colorInternal = 0xFF2020;


// Detect blobs inside the binary image and assign labels to them
// display the results
int numBlobs = BinaryImageOps.labelBlobs4(binary,blobs);
BufferedImage visualBinary = VisualizeBinaryData.renderBinary(binary, false, null);
BufferedImage visualFiltered = VisualizeBinaryData.renderBinary(filtered, false, null);
BufferedImage visualLabel = VisualizeBinaryData.renderLabeledBG(label, contours.size(), null);
BufferedImage visualContour = VisualizeBinaryData.renderContours(contours, colorExternal, colorInternal,
input.width, input.height, null);


// Render the binary image for output and display it in a window
var panel = new ListDisplayPanel();
BufferedImage visualized = VisualizeBinaryData.renderLabeled(blobs, numBlobs, null);
panel.addImage(visualBinary, "Binary Original");
ShowImages.showWindow(visualized,"Labeled Image");
panel.addImage(visualFiltered, "Binary Filtered");
panel.addImage(visualLabel, "Labeled Blobs");
panel.addImage(visualContour, "Contours");
ShowImages.showWindow(panel, "Binary Operations", true);
}
}
}
</syntaxhighlight>
</syntaxhighlight>

Latest revision as of 14:57, 17 January 2022

In a binary image each pixel can have a value of 0 or 1. Binary images are easy to compute and fast to process, which makes them popular in many applications. BoofCV contains many operations for creating and manipulating binary images. The example below demonstrates a few of ones contained inside of BinaryImageOps.

Example File: ExampleBinaryOps.java

Concepts:

  • Image Thresholding
  • Morphological Operations
  • Binary Labeling
  • Pixel Math
  • Image Rendering

Basic Example

/**
 * Demonstrates how to create binary images by thresholding, applying binary morphological operations, and
 * then extracting detected features by finding their contours.
 *
 * @author Peter Abeles
 * @see boofcv.examples.segmentation.ExampleThresholding
 */
public class ExampleBinaryOps {
	public static void main( String[] args ) {
		// load and convert the image into a usable format
		BufferedImage image = UtilImageIO.loadImageNotNull(UtilIO.pathExample("particles01.jpg"));

		// convert into a usable format
		GrayF32 input = ConvertBufferedImage.convertFromSingle(image, null, GrayF32.class);
		var binary = new GrayU8(input.width, input.height);
		var label = new GrayS32(input.width, input.height);

		// Select a global threshold using Otsu's method.
		double threshold = GThresholdImageOps.computeOtsu(input, 0, 255);

		// Apply the threshold to create a binary image
		ThresholdImageOps.threshold(input, binary, (float)threshold, true);

		// remove small blobs through erosion and dilation
		// The null in the input indicates that it should internally declare the work image it needs
		// this is less efficient, but easier to code.
		GrayU8 filtered = BinaryImageOps.erode8(binary, 1, null);
		filtered = BinaryImageOps.dilate8(filtered, 1, null);

		// Detect blobs inside the image using an 8-connect rule
		List<Contour> contours = BinaryImageOps.contour(filtered, ConnectRule.EIGHT, label);

		// colors of contours
		int colorExternal = 0xFFFFFF;
		int colorInternal = 0xFF2020;

		// display the results
		BufferedImage visualBinary = VisualizeBinaryData.renderBinary(binary, false, null);
		BufferedImage visualFiltered = VisualizeBinaryData.renderBinary(filtered, false, null);
		BufferedImage visualLabel = VisualizeBinaryData.renderLabeledBG(label, contours.size(), null);
		BufferedImage visualContour = VisualizeBinaryData.renderContours(contours, colorExternal, colorInternal,
				input.width, input.height, null);

		var panel = new ListDisplayPanel();
		panel.addImage(visualBinary, "Binary Original");
		panel.addImage(visualFiltered, "Binary Filtered");
		panel.addImage(visualLabel, "Labeled Blobs");
		panel.addImage(visualContour, "Contours");
		ShowImages.showWindow(panel, "Binary Operations", true);
	}
}