Abstract: A problem of choosing an appropriate set of parameters for estimates calculating algorithms height optimization is considered. The task arises for example in AVO-polynomial recognition method that depends on height optimization of individual ECA items. Initially some subset of general ECA family was used for such an optimization according to the simplicity of the resulting algorithms. In the present article it is shown that on some general conditions the defined ECA subfamily is optimal meaning that the optimal solution in more general one will belong to the said subfamily.

Keywords: estimates calculating algorithm, optimization, algebraic approach, pattern recognition.

ACM Classification Keywords: I.5 Pattern Recognition — I.5.0 General

Link:

ON CHOOSING PARAMETERS SET FOR ECA HEIGHT OPTIMIZATION

Alexander Dokukin

http://www.foibg.com/ijita/vol22/ijita22-02-p07.pdf