img = img1*mask + img2*(1-mask) How do that ?

原文地址：http://answers.opencv.org/question/160599/img-img1mask-img21-mask-how-do-that/

如何提高一个简单操作的速度？最后的multiply方法很有启示。

Hello,

I would like merge two color images with an image mask.

img1 and img2 are color image with 3 channelsmask is grey image with 1 channel

for merge the two image with the mask I do a loop for each pixel.

float c1,c2;
for(int j = 0; j < img1.rows ; j++ ){
    for(int i = 0; i < img1.cols ; i++ ){
        c1 = (greyGoodScale.at<uchar>(j, i))/255.0;
        c2 = 1-c1;
        img.at<Vec3b>(j, i)[0] = c2*img1.at<Vec3b>(j, i)[0] + c1*img2.at<Vec3b>(j, i)[0];
        img.at<Vec3b>(j, i)[1] = c2*img1.at<Vec3b>(j, i)[1] + c1*img2.at<Vec3b>(j, i)[1];
        img.at<Vec3b>(j, i)[2] = c2*img1.at<Vec3b>(j, i)[2] + c1*img2.at<Vec3b>(j, i)[2];
    }
}

OK, it's work but my image is 720x500 and i have 70ms of processing time is TOO LONG, I need to be real time.I can't do process on GPU.

Is a way to reduce processing time ?

thank.christopheopenCV 3.x

有用的回答：

1

const uchar *scale = &greyGoodScale.at<uchar>(j, 0);
uchar *imgdata = &img.at<uchar>(j, 0);
const uchar *imgdata1 = &img1.at<uchar>(j, 0);
const uchar *imgdata2 = &img2.at<uchar>(j, 0);
for(int i = 0; i < img1.cols ; i++ ){
    c1 = scale[i]/255.0;
    c2 = 1-c1;
    int pos = 3*i;
    for (int k = 0; k < 3; k++){
        imgdata[pos + k] = c2*imgdata1[pos + k] + c1*imgdata2[pos + k];
    }
}

is one a the best solution with 16ms of processing time.

Now I would like divide the loop into 4 threads. My code is ok :

std::thread t1(firstQuarter...
..
std::thread t4(fourthQuarter...
t1.join;
...
t4.join;

but time is exactly same of one thread even time is upper.It's just classic access array memory and for the out image they don't need mutex because each thread have a quarter aof the image. How free array matrix for multithread for this easy case ?

2

    Mat img1 = imread("f:/lib/opencv/samples/data/lena.jpg", IMREAD_COLOR);
    Mat img2 = imread("f:/lib/opencv/samples/data/orange.jpg", IMREAD_COLOR);
    Mat img3,img4;
    Mat mask1(img1.rows, img1.cols, CV_8UC1, Scalar(0));
    circle(mask1, Point(200, 200), 100, Scalar(92), -1);
    Mat mask2=255-mask1;
    Mat mask2c,mask1c;
    vector<Mat> pMask = { mask1,mask1,mask1 };
    merge(pMask, mask1c);
    pMask.clear();

    pMask.push_back(mask2);

    pMask.push_back(mask2);

    pMask.push_back(mask2);
    merge(pMask, mask2c);

    multiply(img1, mask1c, img3, 1.0 / 255, CV_32F);
    multiply(img2, mask2c, img4, 1.0 / 255, CV_32F);
    img4 =img3+img4;
    img4.convertTo(img3,CV_8U);
    imshow("test", img3);
    waitKey();

来自为知笔记(Wiz)