使用OpenCV生成鸟瞰/俯视图。

Question

我正在尝试从一张图片中生成俯视图。对于相机内参和畸变，我使用了从一个安装在车顶上的驾驶模拟器中检索到的硬编码值。

代码的基础来自于《使用OpenCV库学习计算机视觉》第409页。

当我在包含3个内角每行和4个内角每列的棋盘图像上运行代码时，我的鸟瞰图是倒置的。我需要将图像正确转换为鸟瞰图并且朝上，因为我需要用到单应性矩阵进行另一个函数调用。

以下是输入和输出图像以及我正在使用的代码：

#include <highgui.h>
#include <cv.h>
#include <cxcore.h>
#include <math.h>
#include <vector>
#include <stdio.h>

#include <iostream>

using namespace cv;
using namespace std;

int main(int argc, char* argv[]) {

if(argc != 4) return -1;
    // INPUT PARAMETERS:
    //
    int board_w = atoi(argv[1]); //inner corners per row
    int board_h = atoi(argv[2]); //inner corners per column
    int board_n = board_w * board_h;
    CvSize board_sz = cvSize( board_w, board_h );

    //Hard coded intrinsics for the camera  
    Mat intrinsicMat = (Mat_<double>(3, 3) <<
        418.7490, 0., 236.8528,
        0.,558.6650,322.7346,
        0., 0., 1.);      

    //Hard coded distortions for the camera 
    CvMat* distortion = cvCreateMat(1, 4, CV_32F);
    cvmSet(distortion, 0, 0, -0.0019);
    cvmSet(distortion, 0, 1, 0.0161);
    cvmSet(distortion, 0, 2, 0.0011);
    cvmSet(distortion, 0, 3, -0.0016);

    IplImage* image = 0;
    IplImage* gray_image = 0;

    if( (image = cvLoadImage(argv[3])) == 0 ) {
        printf("Error: Couldn’t load %s\n",argv[3]);
        return -1;
    }

    gray_image = cvCreateImage( cvGetSize(image), 8, 1 );
    cvCvtColor(image, gray_image, CV_BGR2GRAY );
    // UNDISTORT OUR IMAGE
    //
    IplImage* mapx = cvCreateImage( cvGetSize(image), IPL_DEPTH_32F, 1 );
    IplImage* mapy = cvCreateImage( cvGetSize(image), IPL_DEPTH_32F, 1 );

    CvMat intrinsic (intrinsicMat);

    //This initializes rectification matrices
    //
    cvInitUndistortMap(
        &intrinsic,
        distortion,
        mapx,
        mapy
    );


    IplImage *t = cvCloneImage(image);
    // Rectify our image
    //

    cvRemap( t, image, mapx, mapy );
    // GET THE CHESSBOARD ON THE PLANE
    //

    cvNamedWindow("Chessboard");
    CvPoint2D32f* corners = new CvPoint2D32f[ board_n ];
    int corner_count = 0;
    int found = cvFindChessboardCorners(
        image,
        board_sz,
        corners,
        &corner_count,
        CV_CALIB_CB_ADAPTIVE_THRESH | CV_CALIB_CB_FILTER_QUADS
    );
    if(!found){
        printf("Couldn’t aquire chessboard on %s, "
        "only found %d of %d corners\n",
        argv[3],corner_count,board_n
    );
    return -1;
    }
    //Get Subpixel accuracy on those corners:
    cvFindCornerSubPix(
        gray_image,
        corners,
        corner_count,
        cvSize(11,11),
        cvSize(-1,-1),
        cvTermCriteria( CV_TERMCRIT_EPS | CV_TERMCRIT_ITER, 30, 0.1 )
    );

    //GET THE IMAGE AND OBJECT POINTS:
    // We will choose chessboard object points as (r,c):
    // (0,0), (board_w-1,0), (0,board_h-1), (board_w-1,board_h-1).
    //

    CvPoint2D32f objPts[4], imgPts[4];
    imgPts[0] = corners[0];
    imgPts[1] = corners[board_w-1];
    imgPts[2] = corners[(board_h-1)*board_w];
    imgPts[3] = corners[(board_h-1)*board_w + board_w-1];

    objPts[0].x = 0; objPts[0].y = 0;
    objPts[1].x = board_w -1; objPts[1].y = 0;
    objPts[2].x = 0; objPts[2].y = board_h -1;
    objPts[3].x = board_w -1; objPts[3].y = board_h -1;


    // DRAW THE POINTS in order: B,G,R,YELLOW
    //
    cvCircle( image, cvPointFrom32f(imgPts[0]), 9, CV_RGB(0,0,255), 3); //blue
    cvCircle( image, cvPointFrom32f(imgPts[1]), 9, CV_RGB(0,255,0), 3); //green
    cvCircle( image, cvPointFrom32f(imgPts[2]), 9, CV_RGB(255,0,0), 3); //red
    cvCircle( image, cvPointFrom32f(imgPts[3]), 9, CV_RGB(255,255,0), 3); //yellow
    // DRAW THE FOUND CHESSBOARD
    //

    cvDrawChessboardCorners(
        image,
        board_sz,
        corners,
        corner_count,
        found
    ); 
    cvShowImage( "Chessboard", image );
    // FIND THE HOMOGRAPHY
    //
    CvMat *H = cvCreateMat( 3, 3, CV_32F);
    cvGetPerspectiveTransform( objPts, imgPts, H);
    Mat homography = H;
    cvSave("Homography.xml",H); //We can reuse H for the same camera mounting

    /**********************GENERATING 3X4 MATRIX***************************/

    // LET THE USER ADJUST THE Z HEIGHT OF THE VIEW
    //
    float Z = 23;
    int key = 0;
    IplImage *birds_image = cvCloneImage(image);
    cvNamedWindow("Birds_Eye");
    // LOOP TO ALLOW USER TO PLAY WITH HEIGHT:
    //
    // escape key stops
    //
    while(key != 27) {
        // Set the height
        //
        CV_MAT_ELEM(*H,float,2,2) = Z;
        // COMPUTE THE FRONTAL PARALLEL OR BIRD’S-EYE VIEW:
        // USING HOMOGRAPHY TO REMAP THE VIEW
        //
    cvWarpPerspective(
    image,
    birds_image,
    H,
    CV_INTER_LINEAR | CV_WARP_INVERSE_MAP | CV_WARP_FILL_OUTLIERS
    );
    cvShowImage( "Birds_Eye", birds_image );
    imwrite("/home/lee/bird.jpg", birds_image);

    key = cvWaitKey();
    if(key == 'u') Z += 0.5;
    if(key == 'd') Z -= 0.5;
    }
    return 0;
}

- Lee Torres

---

你不能简单地将你的单应矩阵乘以180度旋转矩阵吗？或者可以尝试以不同的顺序输入棋盘的角点。 - aardvarkk

注意：图像坐标0,0位于左上角。也许您在两组坐标中使用了底部左侧的原点。 - fmw42

相关（根据一个非答案发布的回答）：https://github.com/neozhaoliang/surround-view-system-introduction/tree/master - starball

Answer 1

经过变换矩阵计算，单应性矩阵的结果看起来是正确的。由于您将相机的z轴映射为世界的y轴，所以鸟瞰图重新映射得到的图像是倒置的。

如果您确实需要将鸟瞰图像作为相机拍摄的图像，可以使用H = Ty * Rx * H的方式进行处理，其中R是绕x轴旋转180度的矩阵，T是在y轴上的平移，而H则是原始单应性矩阵。由于旋转将旧的鸟瞰图重新映射到了y轴的负侧，因此需要进行平移。

- nishitani