All Downloads are FREE. Search and download functionalities are using the official Maven repository.

boofcv.alg.geo.rectify.RectifyFundamental Maven / Gradle / Ivy

Go to download

BoofCV is an open source Java library for real-time computer vision and robotics applications.

There is a newer version: 0.26
Show newest version
/*
 * Copyright (c) 2011-2013, Peter Abeles. All Rights Reserved.
 *
 * This file is part of BoofCV (http://boofcv.org).
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *   http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package boofcv.alg.geo.rectify;

import boofcv.alg.geo.MultiViewOps;
import boofcv.struct.geo.AssociatedPair;
import georegression.geometry.GeometryMath_F64;
import georegression.struct.point.Point2D_F64;
import georegression.struct.point.Point3D_F64;
import georegression.struct.point.Vector3D_F64;
import org.ejml.data.DenseMatrix64F;
import org.ejml.simple.SimpleMatrix;

import java.util.List;

/**
 * 

* Rectifies a stereo pair given a fundamental or essential matrix. The rectification ensures that * the epipolar lines project to infinity along the x-axis. The computed transforms are designed to * minimize the range of disparity between the two images. For this technique to work * the epipoles must lie outside of both images. See [1] for algorithmic details. *

* *

* WARNING: On paper this technique sounds straight forward. In practice it requires a very * precise fundamental matrix estimate to be of practical use. Using the epipolar constraint alone is not * sufficient to remove outliers because a point far away that lands on the epipolar line will have a small * error. Removing lens distortion from the image is recommended. *

* *

* [1] R. Hartley, "Theory and Practice of Projective Rectification", International Journal of Computer Vision, * vol 35, no 2, pages 115-127, 1999. *

* * @author Peter Abeles */ public class RectifyFundamental { // rectifying transform for left and right images private DenseMatrix64F rect1 = new DenseMatrix64F(3,3); private DenseMatrix64F rect2 = new DenseMatrix64F(3,3); // storage for epipoles private Point3D_F64 epipole1 = new Point3D_F64(); private Point3D_F64 epipole2 = new Point3D_F64(); /** * Compute rectification transforms for the stereo pair given a fundamental matrix and its observations. * * @param F Fundamental matrix * @param observations Observations used to compute F * @param width Width of first image. * @param height Height of first image. */ public void process( DenseMatrix64F F , List observations , int width , int height ) { int centerX = width/2; int centerY = height/2; MultiViewOps.extractEpipoles(F, epipole1, epipole2); checkEpipoleInside(width, height); // compute the transform H which will send epipole2 to infinity SimpleMatrix R = rotateEpipole(epipole2,centerX,centerY); SimpleMatrix T = translateToOrigin(centerX,centerY); SimpleMatrix G = computeG(epipole2,centerX,centerY); SimpleMatrix H = G.mult(R).mult(T); //Find the two matching transforms SimpleMatrix Hzero = computeHZero(F,epipole2,H); SimpleMatrix Ha = computeAffineH(observations,H.getMatrix(),Hzero.getMatrix()); rect1.set(Ha.mult(Hzero).getMatrix()); rect2.set(H.getMatrix()); } /** * The epipoles need to be outside the image */ private void checkEpipoleInside(int width, int height) { double x1 = epipole1.x/epipole1.z; double y1 = epipole1.y/epipole1.z; double x2 = epipole2.x/epipole2.z; double y2 = epipole2.y/epipole2.z; if( x1 >= 0 && x1 < width && y1 >= 0 && y1 < height ) throw new IllegalArgumentException("First epipole is inside the image"); if( x2 >= 0 && x2 < width && y2 >= 0 && y2 < height ) throw new IllegalArgumentException("Second epipole is inside the image"); } /** * Create a transform which will move the specified point to the origin */ private SimpleMatrix translateToOrigin( int x0 , int y0 ) { SimpleMatrix T = SimpleMatrix.identity(3); T.set(0, 2, -x0); T.set(1, 2, -y0); return T; } /** * Apply a rotation such that the epipole is equal to [f,0,1)\ */ private SimpleMatrix rotateEpipole( Point3D_F64 epipole , int x0 , int y0 ) { // compute rotation which will set // x * sin(theta) + y * cos(theta) = 0 double x = epipole.x/epipole.z-x0; double y = epipole.y/epipole.z-y0; double theta = Math.atan2(-y,x); double c = Math.cos(theta); double s = Math.sin(theta); SimpleMatrix R = new SimpleMatrix(3,3); R.setRow(0, 0, c,-s); R.setRow(1, 0, s, c); R.set(2, 2, 1); return R; } private SimpleMatrix computeG( Point3D_F64 epipole , int x0 , int y0 ) { double x = epipole.x/epipole.z - x0; double y = epipole.y/epipole.z - y0; double f = Math.sqrt(x*x + y*y); SimpleMatrix G = SimpleMatrix.identity(3); G.set(2,0,-1.0/f); return G; } /** * Finds the values of a,b,c which minimize * * sum (a*x(+)_i + b*y(+)_i + c - x(-)_i)^2 * * See page 306 * * @return Affine transform */ private SimpleMatrix computeAffineH( List observations , DenseMatrix64F H , DenseMatrix64F Hzero ) { SimpleMatrix A = new SimpleMatrix(observations.size(),3); SimpleMatrix b = new SimpleMatrix(A.numRows(),1); Point2D_F64 c = new Point2D_F64(); Point2D_F64 k = new Point2D_F64(); for( int i = 0; i < observations.size(); i++ ) { AssociatedPair a = observations.get(i); GeometryMath_F64.mult(Hzero, a.p1, k); GeometryMath_F64.mult(H,a.p2,c); A.setRow(i,0,k.x,k.y,1); b.set(i,0,c.x); } SimpleMatrix x = A.solve(b); SimpleMatrix Ha = SimpleMatrix.identity(3); Ha.setRow(0,0,x.getMatrix().data); return Ha; } /** * H0 = H*M * P=[M|m] from canonical camera */ private SimpleMatrix computeHZero( DenseMatrix64F F , Point3D_F64 e2 , SimpleMatrix H ) { Vector3D_F64 v = new Vector3D_F64(.1,0.5,.2); // need to make sure M is not singular for this technique to work SimpleMatrix P = SimpleMatrix.wrap(MultiViewOps.canonicalCamera(F, e2, v, 1)); SimpleMatrix M = P.extractMatrix(0, 3, 0, 3); return H.mult(M); } /** * Rectification transform for first camera */ public DenseMatrix64F getRect1() { return rect1; } /** * Rectification transform for second camera */ public DenseMatrix64F getRect2() { return rect2; } }




© 2015 - 2024 Weber Informatics LLC | Privacy Policy