Re: Fixing bug in Harris/Hessian examples

From: Mateusz Loskot (mateusz_at_[hidden])
Date: 2019-09-18 09:50:55

On Mon, 16 Sep 2019 at 19:23, Olzhas Zhumabek
<anonymous.from.applecity_at_[hidden]> wrote:
>
> After thinking about it a bit, it seems like only first and second order
> derivatives are of interest, and higher order ones are very rare. My plan
> at the moment is to add generate_dx_sobel/scharr(unsigned int power) and
> generate_dy_sobel/scharr(unsigned int power) that will generate Sobel and
> Scharr kernels for horizontal and vertical directions respectively.
> [...]
>
> What do you think?

Do you plan those as public functions in boost::gil namespace
or as implementation details in boost::gil::detail namespace?
If the latter, you are free to add any utilities that your solution requires.

My only concerns were on being more careful about new
additions to public interface.

Best regards,
-- 
Mateusz Loskot, http://mateusz.loskot.net