opencv 4.x新增了forEach的方式遍历像素值，比传统方式略快一些。因为它本身是使用多线程并行的方法来遍历的。从opencv源码能看到这句话：

parallel_for_(cv::Range(0, LINES), PixelOperationWrapper(reinterpret_cast<Mat_<_Tp>*>(this), operation));

写了一个测试用例，把它用起来。包括单通道，三通道，浮点型等cv::Mat的遍历。

#include <iostream>

typedef cv::Point3_<uint8_t> Pixel;

void test1(cv::Mat &image)
{
    //raw pointer access.
    double start = (double)cv::getTickCount();
    for (int r = 0; r < image.rows; ++r)
    {
        Pixel *ptr = image.ptr<Pixel>(r, 0);
        const Pixel *ptr_end = ptr + image.cols;
        for (; ptr != ptr_end; ++ptr)
        {
            ptr->x = 255 - ptr->x;
            ptr->y = 255 - ptr->y;
            ptr->z = 255 - ptr->z;
        }
    }

    double time = (((double)cv::getTickCount() - start)) / cv::getTickFrequency();
    printf(" raw pointer access time1 : %.4f seconds\n", time);
}

void test2(cv::Mat &image)
{
    double start = (double)cv::getTickCount();
    int w = image.cols;
    int h = image.rows;
    for (int row = 0; row < h; row++)
    {
        uchar *uc_pixel = image.data + row * image.step;
        for (int col = 0; col < w; col++)
        {
            uc_pixel[0] = 255 - uc_pixel[0];
            uc_pixel[1] = 255 - uc_pixel[1];
            uc_pixel[2] = 255 - uc_pixel[2];
            uc_pixel += 3;
        }
    }

    double time = (((double)cv::getTickCount() - start)) / cv::getTickFrequency();
    printf(" raw pointer access time2 : %.4f seconds\n", time);
}

void test3(cv::Mat &image) //OpenCV中C++11 lambda方式像素遍历,OpenCV4.x开始支持
{
    //forEach方式的像素遍历,三通道图像
    //Pixel和 position(分别指该像素的数值信息和位置信息)
    //使用了x，y，z分别代表该像素点的blue, grean, red这三个通道的颜色值
    //position是遍历的像素点坐标位置
    //position[0]=row, position[1]=col
    double start = (double)cv::getTickCount();
    image.forEach<Pixel>([](Pixel &p, const int *position) -> void {
        p.x = 255 - p.x;
        p.y = 255 - p.y;
        p.z = 255 - p.z;
    });

    double time = (((double)cv::getTickCount() - start)) / cv::getTickFrequency();
    printf(" forEach time3 : %.4f seconds\n", time);
}

void test4(cv::Mat &image)
{
    cv::Mat gray;
    cv::cvtColor(image, gray, cv::COLOR_BGR2GRAY);

    //遍历单通道图像
    //position是遍历的像素点坐标位置
    //position[0]=row, position[1]=col
    gray.forEach<uint8_t>([](uint8_t &p, const int *position) -> void {
        p += 1;
    });
}

void test5(cv::Mat &image)
{
    cv::Mat gray;
    cv::cvtColor(image, gray, cv::COLOR_BGR2GRAY);

    cv::Mat ft;
    gray.convertTo(ft, CV_32FC1, 1.0 / 255.0);

    //多线程并行遍历像素,需要加锁
    float score = 0.8;
    std::vector<cv::Point> vtPos;
    std::vector<float> vtConfidences;
    std::mutex mtx;
    ft.forEach<float>([&vtConfidences, &vtPos, &score, &mtx](float &val, const int *position) -> void {
        if (val > score)
        {
            mtx.lock();
            vtPos.emplace_back(cv::Point(position[1], position[0])); //x,y==col,row
            vtConfidences.emplace_back(val);
            mtx.unlock();
        }
    });

    std::cout << vtPos.size() << std::endl;
}

int main(int argc, char *argv[])
{
    cv::Mat image = cv::imread("D:/temp/2-6-11.jpg", cv::IMREAD_COLOR);
    test1(image);
    test2(image);
    test3(image);
    test4(image);
    test5(image);
    return 0;
}