透視變換

本文就是通過opencv中提供的透視變換函數cv::WarpPerspective(),將左邊的圖像變換為右邊的圖像

原文網址：http://opencv-code.com/tutorials/automatic-perspective-correction-for-quadrilateral-objects/#comment-193

                           

 

具體流程為：

a)載入圖像→灰度化→邊緣處理得到邊緣圖像（edge map）

cv::Mat im = cv::imread(filename);

cv::Mat gray;

cvtColor(im,gray,CV_BGR2GRAY);

Canny(gray,gray,100,150,3);

b)霍夫變換進行直線檢測，此處使用的是probabilistic Hough transform（cv::HoughLinesP）而不是standard Hough transform（cv::HoughLines）

std::vector<Vec4i> lines;

cv::HoughLinesP(gray,lines,1,CV_PI/180,70,30,10);

for(int i = 0; i < lines.size(); i++)

    line(im,cv::Point(lines[i][0],lines[i][1]),cv::Point(lines[i][2],lines[i][3]),Scalar(255,0,0),2,8,0);

c)通過上面的圖我們可以看出，通過霍夫變換檢測到的直線並沒有將整個邊緣包含，但是我們要求的是四個頂點所以並不一定要直線真正的相交，下面就要求四個頂點的坐標，公式為：

 

cv::Point2f computeIntersect(cv::Vec4i a, cv::Vec4i b)
{
    int x1 = a[0], y1 = a[1], x2 = a[2], y2 = a[3];
    int x3 = b[0], y3 = b[1], x4 = b[2], y4 = b[3];

    if (float d = ((float)(x1-x2) * (y3-y4)) - ((y1-y2) * (x3-x4)))
    {
        cv::Point2f pt;
        pt.x = ((x1*y2 - y1*x2) * (x3-x4) - (x1-x2) * (x3*y4 - y3*x4)) / d;
        pt.y = ((x1*y2 - y1*x2) * (y3-y4) - (y1-y2) * (x3*y4 - y3*x4)) / d;
        return pt;
    }
    else
        return cv::Point2f(-1, -1);
}
　　

　　

std::vector<cv::Point2f> corners;
for (int i = 0; i < lines.size(); i++)
{
    for (int j = i+1; j < lines.size(); j++)
    {
        cv::Point2f pt = computeIntersect(lines[i], lines[j]);
        if (pt.x >= 0 && pt.y >= 0)
            corners.push_back(pt);
    }
}

 

d）檢查是不是四邊形

std::vector<cv::Point2f> approx;
cv::approxPolyDP(cv::Mat(corners), approx, 
                 cv::arcLength(cv::Mat(corners), true) * 0.02, true);

if (approx.size() != 4)
{
    std::cout << "The object is not quadrilateral!" << std::endl;
    return -1;
}

　　

 

e)確定四個頂點的具體位置（top-left, bottom-left, top-right, and bottom-right corner）→通過四個頂點求出映射矩陣來.

void sortCorners(std::vector<cv::Point2f>& corners, cv::Point2f center)
{
    std::vector<cv::Point2f> top, bot;

    for (int i = 0; i < corners.size(); i++)
    {
        if (corners[i].y < center.y)
            top.push_back(corners[i]);
        else
            bot.push_back(corners[i]);
    }

    cv::Point2f tl = top[0].x > top[1].x ? top[1] : top[0];
    cv::Point2f tr = top[0].x > top[1].x ? top[0] : top[1];
    cv::Point2f bl = bot[0].x > bot[1].x ? bot[1] : bot[0];
    cv::Point2f br = bot[0].x > bot[1].x ? bot[0] : bot[1];

    corners.clear();
    corners.push_back(tl);
    corners.push_back(tr);
    corners.push_back(br);
    corners.push_back(bl);
}

　下面是獲得中心點坐標然后利用上面的函數確定四個頂點的坐標

for (int i = 0; i < corners.size(); i++)
    center += corners[i];

center *= (1. / corners.size());
sortCorners(corners, center);

　定義目的圖像並初始化為0

cv::Mat quad = cv::Mat::zeros(300, 220, CV_8UC3);

　獲取目的圖像的四個頂點

std::vector<cv::Point2f> dst_pt;
dst.push_back(cv::Point2f(0,0));
dst.push_back(cv::Point2f(quad.cols,0));
dst.push_back(cv::Point2f(quad.cols,quad.rows));
dst.push_back(cv::Point2f(0,quad.rows));

　計算映射矩陣

cv::Mat transmtx = cv::getPerspectiveTransform(corners, quad_pts);

進行透視變換並顯示結果

cv::warpPerspective(im, quad, transmtx, quad.size());
cv::imshow("quadrilateral", quad);