Opencv中at<> 与ptr<> 的不同

Mat M(200, 200, CV_64F);
for(int i = 0; i < M.rows; i++)
{

for(int j = 0; j < M.cols; j++)  
    M.at<double>(i,j)=CV_PI;  

}

/////笔记
Note that the at<> method is not very efficient as it has to calculate the exact
memory position from the pixel row and column. This can be very time consuming
when we process the whole image pixel by pixel. The second method uses the ptr
function, which returns a pointer to a specific image row. The following snippet
obtains the pixel value of each pixel in a color image:
/////节省处理时间的方式，用ptr函数，返回一个指向图像行的指针。
uchar R, G, B;
for (int i = 0; i < src2.rows; i++)
{
Vec3b* pixrow = src2.ptr(i);
for (int j = 0; j < src2.cols; j++)
{
B = pixrow[j][0];
G = pixrow[j][1];
R = pixrow[j][2];
}
}
////笔记
In the example above, ptr is used to get a pointer to the first pixel in each row.
Using this pointer, we can now access each column in the innermost loop.

http://blog.youkuaiyun.com/xuhang0910/article/details/47058419
http://blog.youkuaiyun.com/github_35160620/article/details/51708659
http://blog.youkuaiyun.com/zangle260/article/details/52957063