边缘检测（OpenCV）

梯度算子

 

不变矩

#include <opencv2/highgui.hpp>
#include <opencv2/imgproc.hpp>
#include <iostream>
#include <cmath>

using namespace std;
using namespace cv;

//原图，原图的灰度版，目标图
Mat g_srcImage, g_srcGrayImage, g_dstImage;

//Canny边缘检测相关变量
Mat g_cannyDetectedEdges;
int g_cannyLowThreshold = 1;//TrackBar位置参数  

//Sobel边缘检测相关变量
Mat g_sobelGradient_X, g_sobelGradient_Y;
Mat g_sobelAbsGradient_X, g_sobelAbsGradient_Y;
int g_sobelKernelSize = 1;//TrackBar位置参数  

//Scharr滤波器相关变量
Mat g_scharrGradient_X, g_scharrGradient_Y;
Mat g_scharrAbsGradient_X, g_scharrAbsGradient_Y;

static void ShowHelpText();
static void on_Canny(int, void*);//Canny边缘检测窗口滚动条的回调函数
static void on_Sobel(int, void*);//Sobel边缘检测窗口滚动条的回调函数
void Scharr();// 封装了Scharr边缘检测相关代码的函数
void laplacian(); // 封装了laplacian边缘检测相关代码的函数

/* 不变矩 */
void hu(Mat image);

/* 轮廓 */
Mat src, dst, drawImage;
const char* result = "moments_demo";
int threshold_value = 120;
int threshold_max = 255;
RNG rng(12345);
void Moments_demo(int, void*);
void moments_demo_start();

int main(int argc, char** argv)
{
    ShowHelpText();
    g_srcImage = imread("1.png");
    if (!g_srcImage.data) { printf("Oh，no，读取srcImage错误~！ \n"); return false; }
    namedWindow("【原始图】");
    imshow("【原始图】", g_srcImage);

    g_dstImage.create(g_srcImage.size(), g_srcImage.type());

    cvtColor(g_srcImage, g_srcGrayImage, COLOR_BGR2GRAY);

    namedWindow("【效果图】Canny边缘检测", WINDOW_AUTOSIZE);
    namedWindow("【效果图】Sobel边缘检测", WINDOW_AUTOSIZE);

    createTrackbar("参数值：", "【效果图】Canny边缘检测", &g_cannyLowThreshold, 120, on_Canny);
    createTrackbar("参数值：", "【效果图】Sobel边缘检测", &g_sobelKernelSize, 3, on_Sobel);

    on_Canny(0, 0);
    on_Sobel(0, 0);

    Scharr();
    laplacian();
    // moments_demo_start();

    waitKey();
    return 0;
}

static void ShowHelpText()
{
    printf("当前使用的OpenCV版本为：" CV_VERSION);
    printf("\n运行成功，请调整滚动条观察图像效果\n");
}

void on_Canny(int, void*)
{
    // 先使用 3x3内核来降噪
    blur(g_srcGrayImage, g_cannyDetectedEdges, Size(3, 3));

    // 运行我们的Canny算子
    Canny(g_cannyDetectedEdges, g_cannyDetectedEdges, g_cannyLowThreshold, g_cannyLowThreshold * 3, 3);

    //先将g_dstImage内的所有元素设置为0 
    g_dstImage = Scalar::all(0);

    //使用Canny算子输出的边缘图g_cannyDetectedEdges作为掩码，来将原图g_srcImage拷到目标图g_dstImage中
    g_srcImage.copyTo(g_dstImage, g_cannyDetectedEdges);

    //显示效果图
    imshow("【效果图】Canny边缘检测", g_dstImage);

    //计算边缘图像不变矩
    hu(g_dstImage);
}

void on_Sobel(int, void*)
{
    Sobel(g_srcImage, g_sobelGradient_X, CV_16S, 1, 0, (2 * g_sobelKernelSize + 1), 1, 1, BORDER_DEFAULT);
    convertScaleAbs(g_sobelGradient_X, g_sobelAbsGradient_X);//计算绝对值，并将结果转换成8位
    Sobel(g_srcImage, g_sobelGradient_Y, CV_16S, 0, 1, (2 * g_sobelKernelSize + 1), 1, 1, BORDER_DEFAULT);
    convertScaleAbs(g_sobelGradient_Y, g_sobelAbsGradient_Y);//计算绝对值，并将结果转换成8位
    addWeighted(g_sobelAbsGradient_X, 0.5, g_sobelAbsGradient_Y, 0.5, 0, g_dstImage);
    imshow("【效果图】Sobel边缘检测", g_dstImage);

}

void Scharr() // 
{
    Scharr(g_srcImage, g_scharrGradient_X, CV_16S, 1, 0, 1, 0, BORDER_DEFAULT);
    convertScaleAbs(g_scharrGradient_X, g_scharrAbsGradient_X);//计算绝对值，并将结果转换成8位
    Scharr(g_srcImage, g_scharrGradient_Y, CV_16S, 0, 1, 1, 0, BORDER_DEFAULT);
    convertScaleAbs(g_scharrGradient_Y, g_scharrAbsGradient_Y);//计算绝对值，并将结果转换成8位
    addWeighted(g_scharrAbsGradient_X, 0.5, g_scharrAbsGradient_Y, 0.5, 0, g_dstImage);
    imshow("【效果图】Scharr滤波器", g_dstImage);
}

void laplacian()
{
    Mat result;
    int ksize = 5; // 当ksize==1时，3*3的拉普拉斯算子
    int depth = CV_16S; // 目标图像的深度，当depth==-1时，为原图图像深度
    Laplacian(g_srcImage, g_dstImage, depth, ksize);
    // imshow("【效果图】laplacian滤波器", g_dstImage);
    convertScaleAbs(g_dstImage, result); // (3 + 1) * 0.25
    imshow("Laplacian", result);
}

void Moments_demo(int, void*)
{
    //提取图像边缘
    Mat canny_out;
    Canny(dst, canny_out, threshold_value, threshold_value * 2, 3, false);
    //imshow("canny image", canny_out);

    //发现轮廓，找到图像轮廓
    vector<vector<Point>> contours;
    vector<Vec4i> hierachy;
    findContours(canny_out, contours, hierachy, RETR_TREE, CHAIN_APPROX_SIMPLE, Point(0, 0));

    //计算每个轮廓对象的矩
    vector<    Moments> contours_moments(contours.size());
    vector<Point2f> centers(contours.size());
    for (size_t i = 0; i < contours.size(); i++)
    {
        //计算矩
        contours_moments[i] = moments(contours[i]);
        //moments(InputArray  array,//输入数据
        //bool   binaryImage = false // 是否为二值图像
        centers[i] = Point(static_cast<float>(contours_moments[i].m10 / contours_moments[i].m00), static_cast<float>(contours_moments[i].m01 / contours_moments[i].m00));
        //图像中心Center(x0, y0)=(m10/m00,m01/m00)
    }

    src.copyTo(drawImage);
    for (size_t i = 0; i < contours.size(); i++)
    {
        printf("centers point x：%.2f,y：%.2f\n", centers[i].x, centers[i].y);
        printf("contours %d Area：%.2f Arc length：%.2f \n", i, contourArea(contours[i]), arcLength(contours[i], true));
        Scalar color = Scalar(rng.uniform(0, 255), rng.uniform(0, 255), rng.uniform(0, 255));
        drawContours(drawImage, contours, i, color, 2, LINE_AA, hierachy, 0, Point(0, 0));//绘制轮廓
        circle(drawImage, centers[i], 2, color, 2, LINE_AA);//绘制图形中心
    }
    imshow(result, drawImage);
    return;
}

void moments_demo_start()
{
    src = imread("1.png");
    if (!src.data)
    {
        printf("could not load image...\n");
    }
    char input[] = "gray image";
    namedWindow(input, WINDOW_AUTOSIZE);
    namedWindow(result, WINDOW_AUTOSIZE);
    //输入图像转为灰度图像
    cvtColor(src, dst, COLOR_BGR2GRAY);
    GaussianBlur(dst, dst, Size(3, 3), 0, 0);
    imshow(input, dst);

    const char* thresh = "threshold value";
    createTrackbar(thresh, result, &threshold_value, threshold_max, Moments_demo);
    Moments_demo(0, 0);
}

void hu(Mat image)
{
    cvtColor(image, image, COLOR_BGR2GRAY);
    Moments mts = moments(image);
    double hu[7];
    HuMoments(mts, hu);
    cout << endl << "Canny算子处理后的图像的不变矩 ：" << endl;
    for (int i = 0; i < 7; i++)
    {
        cout << "η" << i+1 << '=' << abs(log(abs(hu[i]))) << endl;
    }
}

 

角点检测

#include <opencv2/opencv.hpp>  
#include <opencv2/imgproc/imgproc.hpp>  
using namespace cv;

int main()
{
    //以灰度模式载入图像并显示
    Mat srcImage = imread("1.jpg", 0);
    imshow("原始图", srcImage);

    //进行Harris角点检测找出角点
    Mat cornerStrength;
    cornerHarris(srcImage, cornerStrength, 2, 3, 0.01);

    //对灰度图进行阈值操作，得到二值图并显示  
    Mat harrisCorner;
    threshold(cornerStrength, harrisCorner, 0.00001, 255, THRESH_BINARY);
    imshow("角点检测后的二值效果图", harrisCorner);

    waitKey(0);
    return 0;
}

 角点检测综合示例

#include <opencv2/opencv.hpp>
#include "opencv2/highgui/highgui.hpp"
#include "opencv2/imgproc/imgproc.hpp"
using namespace cv;
using namespace std;
 
#define WINDOW_NAME1 "【程序窗口1】"        //为窗口标题定义的宏  
#define WINDOW_NAME2 "【程序窗口2】"        //为窗口标题定义的宏  

Mat g_srcImage, g_srcImage1, g_grayImage;
int thresh = 30; //当前阈值
int max_thresh = 175; //最大阈值

void on_CornerHarris(int, void*);//回调函数
static void ShowHelpText();

int main(int argc, char** argv)
{
    ShowHelpText();

    //【1】载入原始图并进行克隆保存
    g_srcImage = imread("1.jpg", 1);
    if (!g_srcImage.data) { printf("读取图片错误，请确定目录下是否有imread函数指定的图片存在~！ \n"); return false; }
    imshow("原始图", g_srcImage);
    g_srcImage1 = g_srcImage.clone();

    //【2】存留一张灰度图
    cvtColor(g_srcImage1, g_grayImage, COLOR_BGR2GRAY);

    //【3】创建窗口和滚动条
    namedWindow(WINDOW_NAME1, WINDOW_AUTOSIZE);
    createTrackbar("阈值: ", WINDOW_NAME1, &thresh, max_thresh, on_CornerHarris);

    //【4】调用一次回调函数，进行初始化
    on_CornerHarris(0, 0);

    waitKey(0);
    return(0);
}

void on_CornerHarris(int, void*)
{
    //---------------------------【1】定义一些局部变量-----------------------------
    Mat dstImage;//目标图
    Mat normImage;//归一化后的图
    Mat scaledImage;//线性变换后的八位无符号整型的图

    //---------------------------【2】初始化---------------------------------------
    //置零当前需要显示的两幅图，即清除上一次调用此函数时他们的值
    dstImage = Mat::zeros(g_srcImage.size(), CV_32FC1);
    g_srcImage1 = g_srcImage.clone();

    //---------------------------【3】正式检测-------------------------------------
    //进行角点检测
    cornerHarris(g_grayImage, dstImage, 2, 3, 0.04, BORDER_DEFAULT);

    // 归一化与转换
    normalize(dstImage, normImage, 0, 255, NORM_MINMAX, CV_32FC1, Mat());
    convertScaleAbs(normImage, scaledImage);//将归一化后的图线性变换成8位无符号整型 

    //---------------------------【4】进行绘制-------------------------------------
    // 将检测到的，且符合阈值条件的角点绘制出来
    for (int j = 0; j < normImage.rows; j++)
    {
        for (int i = 0; i < normImage.cols; i++)
        {
            if ((int)normImage.at<float>(j, i) > thresh + 80)
            {
                circle(g_srcImage1, Point(i, j), 5, Scalar(10, 10, 255), 2, 8, 0);
                circle(scaledImage, Point(i, j), 5, Scalar(0, 10, 255), 2, 8, 0);
            }
        }
    }
    //---------------------------【4】显示最终效果---------------------------------
    imshow(WINDOW_NAME1, g_srcImage1);
    imshow(WINDOW_NAME2, scaledImage);

}

static void ShowHelpText()
{
    printf("当前使用的OpenCV版本为：" CV_VERSION);
    printf("【欢迎来到Harris角点检测示例程序~】");
    printf("\n请调整滚动条观察图像效果");
}

shi_Tomashi角点检测

#include "opencv2/highgui/highgui.hpp"
#include "opencv2/imgproc/imgproc.hpp"
#include <iostream>
using namespace cv;
using namespace std;

#define WINDOW_NAME "【Shi-Tomasi角点检测】"        //为窗口标题定义的宏 

Mat g_srcImage, g_grayImage;
int g_maxCornerNumber = 33;
int g_maxTrackbarNumber = 500;
RNG g_rng(12345);//初始化随机数生成器

void on_GoodFeaturesToTrack(int, void*)
{
    //【1】对变量小于等于1时的处理
    if (g_maxCornerNumber <= 1) { g_maxCornerNumber = 1; }

    //【2】Shi-Tomasi算法（goodFeaturesToTrack函数）的参数准备
    vector<Point2f> corners;
    double qualityLevel = 0.01;//角点检测可接受的最小特征值
    double minDistance = 10;//角点之间的最小距离
    int blockSize = 3;//计算导数自相关矩阵时指定的邻域范围
    double k = 0.04;//权重系数
    Mat copy = g_srcImage.clone();    //复制源图像到一个临时变量中，作为感兴趣区域

    //【3】进行Shi-Tomasi角点检测
    goodFeaturesToTrack(g_grayImage,//输入图像
        corners,//检测到的角点的输出向量
        g_maxCornerNumber,//角点的最大数量
        qualityLevel,//角点检测可接受的最小特征值
        minDistance,//角点之间的最小距离
        Mat(),//感兴趣区域
        blockSize,//计算导数自相关矩阵时指定的邻域范围
        false,//不使用Harris角点检测
        k);//权重系数


    //【4】输出文字信息
    cout << "\t>此次检测到的角点数量为：" << corners.size() << endl;

    //【5】绘制检测到的角点
    int r = 4;
    for (int i = 0; i < corners.size(); i++)
    {
        //以随机的颜色绘制出角点
        circle(copy, corners[i], r, Scalar(g_rng.uniform(0, 255), g_rng.uniform(0, 255),
            g_rng.uniform(0, 255)), -1, 8, 0);
    }

    //【6】显示（更新）窗口
    imshow(WINDOW_NAME, copy);
}


//-----------------------------------【ShowHelpText( )函数】----------------------------------
//          描述：输出一些帮助信息
//----------------------------------------------------------------------------------------------
static void ShowHelpText()
{
    printf("当前使用的OpenCV版本为：" CV_VERSION);
    printf("欢迎来到【Shi-Tomasi角点检测】示例程序\n");
    printf("请调整滑动条观察图像效果\n");

}

int main()
{
    ShowHelpText();

    //【1】载入源图像并将其转换为灰度图
    g_srcImage = imread("1.jpg", 1);
    cvtColor(g_srcImage, g_grayImage, COLOR_BGR2GRAY);

    //【2】创建窗口和滑动条，并进行显示和回调函数初始化
    namedWindow(WINDOW_NAME, WINDOW_AUTOSIZE);
    createTrackbar("最大角点数", WINDOW_NAME, &g_maxCornerNumber, g_maxTrackbarNumber, on_GoodFeaturesToTrack);
    imshow(WINDOW_NAME, g_srcImage);
    on_GoodFeaturesToTrack(0, 0);

    waitKey(0);
    return(0);
}

亚像素级角点检测

#include "opencv2/highgui/highgui.hpp"
#include "opencv2/imgproc/imgproc.hpp"
#include <iostream>
using namespace cv;
using namespace std;

#define WINDOW_NAME "【亚像素级角点检测】"        //为窗口标题定义的宏 

Mat g_srcImage, g_grayImage;
int g_maxCornerNumber = 33;
int g_maxTrackbarNumber = 500;
RNG g_rng(12345);//初始化随机数生成器

void on_GoodFeaturesToTrack(int, void*)
{
    //【1】对变量小于等于1时的处理
    if (g_maxCornerNumber <= 1) { g_maxCornerNumber = 1; }

    //【2】Shi-Tomasi算法（goodFeaturesToTrack函数）的参数准备
    vector<Point2f> corners;
    double qualityLevel = 0.01;//角点检测可接受的最小特征值
    double minDistance = 10;//角点之间的最小距离
    int blockSize = 3;//计算导数自相关矩阵时指定的邻域范围
    double k = 0.04;//权重系数
    Mat copy = g_srcImage.clone();    //复制源图像到一个临时变量中，作为感兴趣区域

    //【3】进行Shi-Tomasi角点检测
    goodFeaturesToTrack(g_grayImage,//输入图像
        corners,//检测到的角点的输出向量
        g_maxCornerNumber,//角点的最大数量
        qualityLevel,//角点检测可接受的最小特征值
        minDistance,//角点之间的最小距离
        Mat(),//感兴趣区域
        blockSize,//计算导数自相关矩阵时指定的邻域范围
        false,//不使用Harris角点检测
        k);//权重系数

    //【4】输出文字信息
    cout << "\n\t>-------------此次检测到的角点数量为：" << corners.size() << endl;

    //【5】绘制检测到的角点
    int r = 4;
    for (unsigned int i = 0; i < corners.size(); i++)
    {
        //以随机的颜色绘制出角点
        circle(copy, corners[i], r, Scalar(g_rng.uniform(0, 255), g_rng.uniform(0, 255),
            g_rng.uniform(0, 255)), -1, 8, 0);
    }

    //【6】显示（更新）窗口
    imshow(WINDOW_NAME, copy);

    //【7】亚像素角点检测的参数设置
    Size winSize = Size(5, 5);
    Size zeroZone = Size(-1, -1);
    //此句代码的OpenCV2版为：
    //TermCriteria criteria = TermCriteria( CV_TERMCRIT_EPS + CV_TERMCRIT_ITER, 40, 0.001 );
    //此句代码的OpenCV3版为：
    TermCriteria criteria = TermCriteria(TermCriteria::EPS + TermCriteria::MAX_ITER, 40, 0.001);

    //【8】计算出亚像素角点位置
    cornerSubPix(g_grayImage, corners, winSize, zeroZone, criteria);

    //【9】输出角点信息
    for (int i = 0; i < corners.size(); i++)
    {
        cout << " \t>>精确角点坐标[" << i << "]  (" << corners[i].x << "," << corners[i].y << ")" << endl;
    }


}

static void ShowHelpText()
{
    printf("当前使用的OpenCV版本为：" CV_VERSION);
    printf("欢迎来到【亚像素级角点检测】示例程序\n");
    printf("请调整滑动条观察图像效果\n");

}

int main()
{
    ShowHelpText();

    //【1】载入源图像并将其转换为灰度图
    g_srcImage = imread("1.jpg", 1);
    cvtColor(g_srcImage, g_grayImage, COLOR_BGR2GRAY);

    //【2】创建窗口和滑动条，并进行显示和回调函数初始化
    namedWindow(WINDOW_NAME, WINDOW_AUTOSIZE);
    createTrackbar("最大角点数", WINDOW_NAME, &g_maxCornerNumber, g_maxTrackbarNumber, on_GoodFeaturesToTrack);
    imshow(WINDOW_NAME, g_srcImage);
    on_GoodFeaturesToTrack(0, 0);

    waitKey(0);
    return(0);
}

 

---------------------continue------------------------------------------