Abstract

This paper proposes robustness indices for an algorithm for recognizing the type of a dynamic object and a probability-distribution law for the sufficient recognition statistics formed by the algorithm when there is a priori indeterminacy. The law is used to validate the algorithm. The wavelet–fractal-correlation algorithm implements a vector criterion of the ratios of the likelihood functions of simple alternative hypotheses—the types of objects invariant to the features of their trajectories. The likelihood functions are reconstituted by modeling over assemblages of implementations of fractal dimensions, energies of the wavelet spectra, and the maximum eigenvalues of displaced correlation matrices as functionals of the coordinates of the spatial position of various types of actual objects measured by an optoelectronic device. Modeling is used to confirm that the algorithm is stable and highly efficient.

© 2018 Optical Society of America

PDF Article

References

You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription