Аннотация:The methods of Gradient Morphology (GM) are based on the Finite Dimensional Sampling Theorem (FDST), which allows us to calculate partial derivatives of arrays of numbers, almost with the accuracy of the mantissa (without the use of finite difference schemes), in particular, to calculate gradient fields B=grad P from scalar images P.
Methods for analyzing the shape of vector fields B allow GM methods in images P to accurately estimate the parameters of vortex formation, for example or parameters of a ship's wake structure in an image.