Similar to std::transform function for cv::Mat in OpenCV

Question

I'm wondering is there any function in OpenCV, which operates with two cv::Mat objects alike std::tranform in STL?

For example,

cv::transform(src1, src2, dst, bin_function);

where

• src1 and src2 : input matrices (cv::Mat) of the same size.
• dst : output matrix of the same as inputs.
• bin_function : binary function or function object that accepts two elements as argument (one of each of the two matrices), and returns some result value.

I tried to search in documentation but not found such thing.

Answer 1

As far as I know there is no such function in OpenCV yet. However, you can use the STL algorithm std::transform on the cv::Mat iterators like this:

std::transform( src1.begin<float>(), src1.end<float>(), 
                src2.begin<float>(), dest.begin<float>(), bin_function );

Be sure to use the right template type argument. So if you have a 3-channel matrices with 8 bit for each channel for example, then use Vec3b instead of float .

Beware of the usual STL pitfalls though. src1 , src2 and dest should have the same number of elements before your call to std::transform . If you need this genericity, you may consider to write a wrapper function cv::transform yourself, which checks that, reserves memory for dest . Finding the right type for the template argument dynamically will probably be difficult, since the bin_function needs to accept all possible arguments. However, even that may be possible in certain specific cases, if you use C++14 lambdas with auto parameters.