Invariant pattern recognition by moment fourier descriptor

摘要

This paper proposes a new shape descriptor, called moment Fourier descriptor, which can describe a complex object composed by a set of closed regions. This descriptor is shown to be independent of object's translation, rotation and scaling. The essential advantage of moment Fourier descriptor is that it can be used to recognize more complex patterns than the traditional Fourier descriptors. Simulations are given to show the high performance of the proposed approach.