Example Rectification Calibrated - Revision history

Peter at 03:31, 9 October 2021

2021-10-09T03:31:39Z

← Older revision		Revision as of 03:31, 9 October 2021
Line 8:		Line 8:
	Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.		Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.

	Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.38/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]		Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.39/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]

	Concepts:		Concepts:
Line 26:		Line 26:
	*		*
	* The example code does the following:		* The example code does the following:
	* <lo>		* <ol>
	* <li>Load stereo extrinsic and intrinsic parameters from a file along with a pair of images.</li>		* <li>Load stereo extrinsic and intrinsic parameters from a file along with a pair of images.</li>
	* <li>Undistort and rectify images. This provides one rectification matrix for each image along with a new		* <li>Undistort and rectify images. This provides one rectification matrix for each image along with a new
Line 32:		Line 32:
	* <li>The original rectification does not try to maximize view area, however it can be adjusted.</li>		* <li>The original rectification does not try to maximize view area, however it can be adjusted.</li>
	* <li>After rectification is finished the results are displayed.</li>		* <li>After rectification is finished the results are displayed.</li>
	* </lo>		* </ol>
	*		*
	* Note that the y-axis in left and right images align after rectification. You can click in the images to draw a line		* Note that the y-axis in left and right images align after rectification. You can click in the images to draw a line
Line 81:		Line 81:

	// undistorted and rectify images		// undistorted and rectify images
	~~FMatrixRMaj~~ rect1_F32 = new FMatrixRMaj(3, 3); // TODO simplify code some how		var rect1_F32 = new FMatrixRMaj(3, 3); // TODO simplify code some how
	~~FMatrixRMaj~~ rect2_F32 = new FMatrixRMaj(3, 3);		var rect2_F32 = new FMatrixRMaj(3, 3);
	ConvertMatrixData.convert(rect1, rect1_F32);		ConvertMatrixData.convert(rect1, rect1_F32);
	ConvertMatrixData.convert(rect2, rect2_F32);		ConvertMatrixData.convert(rect2, rect2_F32);

	ImageDistort rectifyImageLeft =		ImageDistort<Planar<GrayF32>, Planar<GrayF32>> rectifyImageLeft =
	RectifyDistortImageOps.rectifyImage(param.getLeft(), rect1_F32, BorderType.SKIP, distLeft.getImageType());		RectifyDistortImageOps.rectifyImage(param.getLeft(), rect1_F32, BorderType.SKIP, distLeft.getImageType());
	ImageDistort rectifyImageRight =		ImageDistort<Planar<GrayF32>, Planar<GrayF32>> rectifyImageRight =
	RectifyDistortImageOps.rectifyImage(param.getRight(), rect2_F32, BorderType.SKIP, distRight.getImageType());		RectifyDistortImageOps.rectifyImage(param.getRight(), rect2_F32, BorderType.SKIP, distRight.getImageType());

Line 99:		Line 99:

	// show results and draw a horizontal line where the user clicks to see rectification easier		// show results and draw a horizontal line where the user clicks to see rectification easier
	~~ListDisplayPanel~~ panel = new ListDisplayPanel();		var panel = new ListDisplayPanel();
	panel.addItem(new RectifiedPairPanel(true, origLeft, origRight), "Original");		panel.addItem(new RectifiedPairPanel(true, origLeft, origRight), "Original");
	panel.addItem(new RectifiedPairPanel(true, outLeft, outRight), "Rectified");		panel.addItem(new RectifiedPairPanel(true, outLeft, outRight), "Rectified");

Peter at 20:10, 12 July 2021

2021-07-12T20:10:46Z

← Older revision		Revision as of 20:10, 12 July 2021
Line 8:		Line 8:
	Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.		Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.

	Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.37/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]		Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.38/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]

	Concepts:		Concepts:

Peter at 02:48, 22 December 2020

2020-12-22T02:48:47Z

← Older revision		Revision as of 02:48, 22 December 2020
Line 8:		Line 8:
	Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.		Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.

	Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.36/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]		Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.37/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]

	Concepts:		Concepts:
Line 21:		Line 21:
	<syntaxhighlight lang="java">		<syntaxhighlight lang="java">
	/**		/**
	* <p>		* Shows how to rectify a pair of stereo images with known intrinsic parameters and stereo baseline. When you
	* Shows how to rectify a pair of stereo images with known intrinsic parameters and stereo baseline.		* rectify a stereo pair you are applying a transform that removes lens distortion and "rotates" the views
	* ~~The example code does the following:<br>~~		* such that they are parallel to each other, facilitating stereo processing.
	* 1) Load stereo ~~parameters from XML file with~~ a ~~pair of images.<br>~~
	* 2) Undistort and ~~rectify images.. This provides one rectification matrix~~
	* ~~for~~ each ~~image along with a new camera calibration matrix.<br>~~
	* 3) The original rectification does not try to maximize view area, ~~however it can be adjusted~~.
	* 4)After rectification is finished the results are displayed.<br>
	* </p>
	*		*
	* <p>		* The example code does the following:
	* Note that the y-axis in left and right images align after rectification. The curved image ~~edge~~		* <lo>
	* is an artifact of lens distortion being removed.		* <li>Load stereo extrinsic and intrinsic parameters from a file along with a pair of images.</li>
	* </p>		* <li>Undistort and rectify images. This provides one rectification matrix for each image along with a new
			* camera calibration matrix.</li>
			* <li>The original rectification does not try to maximize view area, however it can be adjusted.</li>
			* <li>After rectification is finished the results are displayed.</li>
			* </lo>
			*
			* Note that the y-axis in left and right images align after rectification. You can click in the images to draw a line
			* that makes this easy to see. The curved image birder is an artifact of lens distortion being removed.
	*		*
	* @author Peter Abeles		* @author Peter Abeles
	*/		*/
	public class ExampleRectifyCalibratedStereo {		public class ExampleRectifyCalibratedStereo {
			public static void main( String[] args ) {
	public static void main( String ~~args~~[] ) {
	String dir = UtilIO.pathExample("calibration/stereo/Bumblebee2_Chess/");		String dir = UtilIO.pathExample("calibration/stereo/Bumblebee2_Chess/");

	StereoParameters param = CalibrationIO.load(new File(dir , "stereo.yaml"));		StereoParameters param = CalibrationIO.load(new File(dir, "stereo.yaml"));

	// load images		// load images
	BufferedImage origLeft = UtilImageIO.loadImage(dir,"left05.jpg");		BufferedImage origLeft = UtilImageIO.loadImage(dir, "left05.jpg");
	BufferedImage origRight = UtilImageIO.loadImage(dir,"right05.jpg");		BufferedImage origRight = UtilImageIO.loadImage(dir, "right05.jpg");

	// distorted images		// distorted images
	Planar<GrayF32> distLeft =		Planar<GrayF32> distLeft =
	ConvertBufferedImage.convertFromPlanar(origLeft, null,true, GrayF32.class);		ConvertBufferedImage.convertFromPlanar(origLeft, null, true, GrayF32.class);
	Planar<GrayF32> distRight =		Planar<GrayF32> distRight =
	ConvertBufferedImage.convertFromPlanar(origRight, null,true, GrayF32.class);		ConvertBufferedImage.convertFromPlanar(origRight, null, true, GrayF32.class);

	// storage for undistorted + rectified images		// storage for undistorted + rectified images
Line 67:		Line 67:
	DMatrixRMaj K2 = PerspectiveOps.pinholeToMatrix(param.getRight(), (DMatrixRMaj)null);		DMatrixRMaj K2 = PerspectiveOps.pinholeToMatrix(param.getRight(), (DMatrixRMaj)null);

	rectifyAlg.process(K1,new Se3_F64(),K2,leftToRight);		rectifyAlg.process(K1, new Se3_F64(), K2, leftToRight);

	// rectification matrix for each image		// rectification matrix for each image
	DMatrixRMaj rect1 = rectifyAlg.~~getRect1~~();		DMatrixRMaj rect1 = rectifyAlg.getUndistToRectPixels1();
	DMatrixRMaj rect2 = rectifyAlg.~~getRect2~~();		DMatrixRMaj rect2 = rectifyAlg.getUndistToRectPixels2();
	// New calibration matrix,		// New calibration matrix,
	// Both cameras have the same one after rectification.		// Both cameras have the same one after rectification.
Line 81:		Line 81:

	// undistorted and rectify images		// undistorted and rectify images
	FMatrixRMaj rect1_F32 = new FMatrixRMaj(3,3); // TODO simplify code some how		FMatrixRMaj rect1_F32 = new FMatrixRMaj(3, 3); // TODO simplify code some how
	FMatrixRMaj rect2_F32 = new FMatrixRMaj(3,3);		FMatrixRMaj rect2_F32 = new FMatrixRMaj(3, 3);
	ConvertMatrixData.convert(rect1, rect1_F32);		ConvertMatrixData.convert(rect1, rect1_F32);
	ConvertMatrixData.convert(rect2, rect2_F32);		ConvertMatrixData.convert(rect2, rect2_F32);

	ImageDistort rectifyImageLeft =		ImageDistort rectifyImageLeft =
	~~RectifyImageOps~~.rectifyImage(param.getLeft(), rect1_F32, BorderType.SKIP, distLeft.getImageType());		RectifyDistortImageOps.rectifyImage(param.getLeft(), rect1_F32, BorderType.SKIP, distLeft.getImageType());
	ImageDistort rectifyImageRight =		ImageDistort rectifyImageRight =
	~~RectifyImageOps~~.rectifyImage(param.getRight(), rect2_F32, BorderType.SKIP, distRight.getImageType());		RectifyDistortImageOps.rectifyImage(param.getRight(), rect2_F32, BorderType.SKIP, distRight.getImageType());

	rectifyImageLeft.apply(distLeft,rectLeft);		rectifyImageLeft.apply(distLeft, rectLeft);
	rectifyImageRight.apply(distRight,rectRight);		rectifyImageRight.apply(distRight, rectRight);

	// convert for output		// convert for output
	BufferedImage outLeft = ConvertBufferedImage.convertTo(rectLeft,null,true);		BufferedImage outLeft = ConvertBufferedImage.convertTo(rectLeft, null, true);
	BufferedImage outRight = ConvertBufferedImage.convertTo(rectRight, null,true);		BufferedImage outRight = ConvertBufferedImage.convertTo(rectRight, null, true);

	// show results and draw a horizontal line where the user clicks to see rectification easier		// show results and draw a horizontal line where the user clicks to see rectification easier
Line 103:		Line 103:
	panel.addItem(new RectifiedPairPanel(true, outLeft, outRight), "Rectified");		panel.addItem(new RectifiedPairPanel(true, outLeft, outRight), "Rectified");

	ShowImages.showWindow(panel,"Stereo Rectification Calibrated",true);		ShowImages.showWindow(panel, "Stereo Rectification Calibrated", true);
	}		}
	}		}
	</syntaxhighlight>		</syntaxhighlight>

Peter at 05:32, 18 May 2020

2020-05-18T05:32:56Z

← Older revision		Revision as of 05:32, 18 May 2020
Line 8:		Line 8:
	Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.		Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.

	Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.35/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]		Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.36/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]

	Concepts:		Concepts:
Line 77:		Line 77:

	// Adjust the rectification to make the view area more useful		// Adjust the rectification to make the view area more useful
	RectifyImageOps.fullViewLeft(param.left~~, null~~, rect1, rect2, rectK, null);		RectifyImageOps.fullViewLeft(param.left, rect1, rect2, rectK, null);
	// RectifyImageOps.allInsideLeft(param.left~~, leftHanded~~, rect1, rect2, rectK);		// RectifyImageOps.allInsideLeft(param.left, rect1, rect2, rectK, null);

	// undistorted and rectify images		// undistorted and rectify images

Peter at 01:04, 24 December 2019

2019-12-24T01:04:35Z

← Older revision		Revision as of 01:04, 24 December 2019
Line 8:		Line 8:
	Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.		Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.

	Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.33/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]		Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.35/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]

	Concepts:		Concepts:
Line 77:		Line 77:

	// Adjust the rectification to make the view area more useful		// Adjust the rectification to make the view area more useful
	RectifyImageOps.fullViewLeft(param.left, rect1, rect2, rectK);		RectifyImageOps.fullViewLeft(param.left, null, rect1, rect2, rectK, null);
	// RectifyImageOps.allInsideLeft(param.left, leftHanded, rect1, rect2, rectK);		// RectifyImageOps.allInsideLeft(param.left, leftHanded, rect1, rect2, rectK);

Peter at 15:04, 15 March 2019

2019-03-15T15:04:14Z

← Older revision		Revision as of 15:04, 15 March 2019
Line 8:		Line 8:
	Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.		Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.

	Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.32/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]		Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.33/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]

	Concepts:		Concepts:
Line 15:		Line 15:

	Related Examples:		Related Examples:
	* [[Example_Rectification_Uncalibrated\| Uncalibrated Stereo Rectification]]
	* [[Example_Calibrate_Planar_Stereo\| Stereo Camera Calibration]]		* [[Example_Calibrate_Planar_Stereo\| Stereo Camera Calibration]]
	* [[Example_Remove_Lens_Distortion\| Removing Lens Distortion]]		* [[Example_Remove_Lens_Distortion\| Removing Lens Distortion]]

Peter at 04:57, 27 December 2018

2018-12-27T04:57:51Z

← Older revision		Revision as of 04:57, 27 December 2018
Line 6:		Line 6:
	</center>		</center>

	Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.		Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.

	Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.32/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]		Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.32/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]

Peter at 04:06, 27 December 2018

2018-12-27T04:06:13Z

← Older revision		Revision as of 04:06, 27 December 2018
Line 8:		Line 8:
	Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.		Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.

	Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.31/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]		Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.32/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]

	Concepts:		Concepts:

Peter at 03:06, 17 October 2018

2018-10-17T03:06:16Z

← Older revision		Revision as of 03:06, 17 October 2018
Line 8:		Line 8:
	Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.		Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.

	Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.27/examples/src/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]		Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.31/examples/src/main/java/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]

	Concepts:		Concepts:
Line 65:		Line 65:

	// original camera calibration matrices		// original camera calibration matrices
	DMatrixRMaj K1 = PerspectiveOps.~~calibrationMatrix~~(param.getLeft(), (DMatrixRMaj)null);		DMatrixRMaj K1 = PerspectiveOps.pinholeToMatrix(param.getLeft(), (DMatrixRMaj)null);
	DMatrixRMaj K2 = PerspectiveOps.~~calibrationMatrix~~(param.getRight(), (DMatrixRMaj)null);		DMatrixRMaj K2 = PerspectiveOps.pinholeToMatrix(param.getRight(), (DMatrixRMaj)null);

	rectifyAlg.process(K1,new Se3_F64(),K2,leftToRight);		rectifyAlg.process(K1,new Se3_F64(),K2,leftToRight);

Peter at 16:15, 17 August 2017

2017-08-17T16:15:08Z

← Older revision		Revision as of 16:15, 17 August 2017
Line 8:		Line 8:
	Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.		Stereo rectification is the process of distorting an image such that the epipoles of both images are at infinity. If the epipoles are at infinity along the x-axis, then corresponding features must lie along the same y-coordinates. Using knowledge that correspondence feature's have the same y-coordinate allows for quick searches. Many stereo vision algorithm rely on rectification. The example below demonstrates rectification for a calibrated stereo pair. Note that after calibration the new camera view has a different set of intrinsic parameters.

	Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.25/examples/src/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]		Example File: [https://github.com/lessthanoptimal/BoofCV/blob/v0.27/examples/src/boofcv/examples/stereo/ExampleRectifyCalibratedStereo.java ExampleRectifyCalibratedStereo.java]

	Concepts:		Concepts:
Line 52:		Line 52:
	// distorted images		// distorted images
	Planar<GrayF32> distLeft =		Planar<GrayF32> distLeft =
	ConvertBufferedImage.~~convertFromMulti~~(origLeft, null,true, GrayF32.class);		ConvertBufferedImage.convertFromPlanar(origLeft, null,true, GrayF32.class);
	Planar<GrayF32> distRight =		Planar<GrayF32> distRight =
	ConvertBufferedImage.~~convertFromMulti~~(origRight, null,true, GrayF32.class);		ConvertBufferedImage.convertFromPlanar(origRight, null,true, GrayF32.class);

	// storage for undistorted + rectified images		// storage for undistorted + rectified images
Line 65:		Line 65:

	// original camera calibration matrices		// original camera calibration matrices
	~~DenseMatrix64F~~ K1 = PerspectiveOps.calibrationMatrix(param.getLeft(), null);		DMatrixRMaj K1 = PerspectiveOps.calibrationMatrix(param.getLeft(), (DMatrixRMaj)null);
	~~DenseMatrix64F~~ K2 = PerspectiveOps.calibrationMatrix(param.getRight(), null);		DMatrixRMaj K2 = PerspectiveOps.calibrationMatrix(param.getRight(), (DMatrixRMaj)null);

	rectifyAlg.process(K1,new Se3_F64(),K2,leftToRight);		rectifyAlg.process(K1,new Se3_F64(),K2,leftToRight);

	// rectification matrix for each image		// rectification matrix for each image
	~~DenseMatrix64F~~ rect1 = rectifyAlg.getRect1();		DMatrixRMaj rect1 = rectifyAlg.getRect1();
	~~DenseMatrix64F~~ rect2 = rectifyAlg.getRect2();		DMatrixRMaj rect2 = rectifyAlg.getRect2();
	// New calibration matrix,		// New calibration matrix,
	// Both cameras have the same one after rectification.		// Both cameras have the same one after rectification.
	~~DenseMatrix64F~~ rectK = rectifyAlg.getCalibrationMatrix();		DMatrixRMaj rectK = rectifyAlg.getCalibrationMatrix();

	// Adjust the rectification to make the view area more useful		// Adjust the rectification to make the view area more useful
Line 82:		Line 82:

	// undistorted and rectify images		// undistorted and rectify images
			FMatrixRMaj rect1_F32 = new FMatrixRMaj(3,3); // TODO simplify code some how
			FMatrixRMaj rect2_F32 = new FMatrixRMaj(3,3);
			ConvertMatrixData.convert(rect1, rect1_F32);
			ConvertMatrixData.convert(rect2, rect2_F32);

	ImageDistort rectifyImageLeft =		ImageDistort rectifyImageLeft =
	RectifyImageOps.rectifyImage(param.getLeft(), ~~rect1~~, BorderType.SKIP, distLeft.getImageType());		RectifyImageOps.rectifyImage(param.getLeft(), rect1_F32, BorderType.SKIP, distLeft.getImageType());
	ImageDistort rectifyImageRight =		ImageDistort rectifyImageRight =
	RectifyImageOps.rectifyImage(param.getRight(), ~~rect2~~, BorderType.SKIP, distRight.getImageType());		RectifyImageOps.rectifyImage(param.getRight(), rect2_F32, BorderType.SKIP, distRight.getImageType());

	rectifyImageLeft.apply(distLeft,rectLeft);		rectifyImageLeft.apply(distLeft,rectLeft);