Accessing each pixel of a Mat

Question

Today I stumbled across this (for me) weird thing, namely when I run this code

int i,j;
uint8_t* p = y.data;
for( i = 0; i < y.rows; ++i)
{
    for ( j = 0; j < y.cols; ++j)
    {
       std::cout << y.at<double>(i,j) << std::endl;
       std::cout << saturate_cast<double>(p[i*y.cols + j]) << std::endl;
    }
}

the output I get looks like this:

0.00683212
251
0.00683212
123
0.00683212
63
0.00683212
254
0.00683212
251
0.00683212
123
...

can someone explain to me why is it so?

Shouldn't it be same?

Answer 1

Here's what it looks like the memory pointed at by p is:

// |251 |123 |63  |254 |...   uint8_t
// +----+----+----+----+----+
// | FB | 7B | 3F | FE |... | hex
// +----+----+----+----+----+
// |0.00683212          ...   double

when you do y.at<double>(i,j) what's essentially going on is:

p[i*y.cols + j]

you are accessing the array as a uint8_t and therefore get a uint8_t which converted to double keeps the same value. When you access it through .at<double>() , its accessing it as a double*