Example Fiducial Square Image

From BoofCV
Revision as of 16:37, 15 September 2014 by Peter (talk | contribs) (Created page with "<center> <gallery widths=280px heights=240px> file:Example_fiducial_image.jpg | Rendered 3D flat squares on top of fiducials </gallery> </center> Demonstration how to detect...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Demonstration how to detect square image fiducials. After the fiducial detector has been created a description of each image it detects is passed in. These images are converted into binary images and resized if needed. A large number of unique fiducials can be detected with a linear growth in computational time.

Example Code:


  • Fiducials
  • Pose estimation

Relevant Tutorials:

Relevant Applets:

Example Code

 * Detects square binary fiducials inside an image, writes out there pose, and visualizes a virtual flat cube
 * above them in the input image.
 * @author Peter Abeles
public class ExampleFiducialImage {
	public static void main(String[] args) {

		String directory = "../data/applet/fiducial/image/";

		String imageName = "image0000.jpg";
//		String imageName = "image0001.jpg";
//		String imageName = "image0002.jpg";

		// load the lens distortion parameters and the input image
		IntrinsicParameters param = UtilIO.loadXML(directory + "intrinsic.xml");
		BufferedImage input = UtilImageIO.loadImage(directory + imageName);
		ImageFloat32 original = ConvertBufferedImage.convertFrom(input, true, ImageType.single(ImageFloat32.class));

		// Detect the fiducial
		SquareImage_to_FiducialDetector<ImageFloat32> detector = FactoryFiducial.
				squareImageRobust(new ConfigFiducialImage(0.1), 6, ImageFloat32.class);
//				squareImageFast(new ConfigFiducialImage(0.1), 100, ImageFloat32.class);

		// give it a description of all the targets
		ImageFloat32 dog = UtilImageIO.loadImage(directory + "dog.png",ImageFloat32.class);
		detector.addTarget(dog, 125);
		// uncomment to detect the text target
		ImageFloat32 text = UtilImageIO.loadImage(directory + "text.png",ImageFloat32.class);
		detector.addTarget(text, 125);



		// print the results
		Graphics2D g2 = input.createGraphics();
		Se3_F64 targetToSensor = new Se3_F64();
		for (int i = 0; i < detector.totalFound(); i++) {
			System.out.println("Target ID = "+detector.getId(i));
			detector.getFiducialToWorld(i, targetToSensor);