Knowing-How on Boundary Geometric Moments

Abstract

In this research the perfomance of the Chen's Improved (Boundary) Moments is carefully compared to tha of the traditional (Masive) Moments to achieve this investigation, the pattern recognition power of the former is thoroughly assessed against that of the latter. The boundary moments are evaluated by two methods, in the first by edge-tracing, in the second methods, the edge pixels are considered as they are met when sweeping the image space. It is found that the average “distance” between a reference and trial samples for massive and boundary moments yield approximate values, implying that these two methods are equivalent concerning accurateness.

It is concluded that the computation of the Boundary Moments by sweeping the image space associates minimun computational complexity to a high enough object classification efficiency, thus they may be used in lieu of the traditional moments.

The research includes hollowed objects, it is experimentally demonstrated that pattern classification of this kind of objects can also be successfully achieved whit the boundary moments, provided that they are evaluated by sweeping the image space.