Do anyone knows why medianBlur use a KxK size for the Matrix Kernel while Blur and Gaussian use a MxN matrix as Kernel? May be you @stevenputtemans know the answer ;)
↧