On the Use of Non-Orthogonal Signal Expansions for Recognitlon

This paper presents a novel approah for template recognition by signal expansion into a set of non-orthognal template-similar basis functions. Expansion matching is a special case of the general non-orthogonal expansion and is equivalent to "restoration" of undegraded signals. It is quite robust in conditions of noise, superposition and severe occultation. It is shown that expansion matching maximizes a new and more practically defined Discriminative signal-to-noise ratio (DSNR). Theoretical and experimental comparisons with the widely used correlation matching (matched filtering), show that expansion matching yields much higher DSNR.