Olzhas Zhumabek

@sdebionne @lpranam @mloskot I believe we could mount a lot of functionality on properly written matrix type. Here is the first sketch: https://github.com/simmplecoder/gil/blob/2D_Convolution_and_Correlation/include/boost/gil/extension/numeric/arithmetics.hpp https://github.com/simmplecoder/gil/blob/2D_Convolution_and_Correlation/include/boost/gil/extension/numeric/matrix.hpp These are very rough sketches of...

@meshtag there is a way to improve performance by using associativity property of convolution. Lets say you have some base kernel `k` and you want to raise it to power...

@meshtag I believe it should also be possible to use some container like `std::vector`, use `gil::interleaved_view` or similar, use already existing convolution on image views, then just put the resulting...

I believe there is a bug in the last hysteresis step: it will only detect Г shaped edges, and not horizontally flipped L shaped edges. Let me give an example...

Sorry for choosing confusing wording. I meant that weak edges that should've been included in the output will not be included. Thanks for clarification on the decisions.

I had a look at the error GA are giving. It seems like the compiler uses C++11 mode but standard library doesn't ... It seems more like C++20 is being...

I was thinking about making the argument a template, after adding `boost::gil::gray64f_view_t` and tests. Casting after making a template might suppress warnings that would otherwise be useful. Do other Boost...

@Neumann-A I will try to get a commit in this week, next week it should be implemented. Do you have some easy test case I could throw into the commit...

Sorry last days at work were very hectic. I have created a draft PR https://github.com/boostorg/gil/pull/743, I am a bit unsure how to create a test for it yet, but after...

I will give it a go then, should have something to show next week or so.