Pattern recognition with variable discrimination capability by dual non-linear optical correlation

The discrimination capability requirements of pattern recognition systems m ay vary from one given purpose to another. In this work a recognition syste m with variable and selective discrimination capability is obtained by appl ying a dual non-linear correlation (DNC) model to a joint-transform correla tor. DNC is obtained by means of two non-linear operators that are applied to both the reference and input channels. A particular DNC is given by the values taken by two real control parameters that determine the non-linear o perators. In comparison with conventional filtering methods, an increased a nd variable discrimination capability is achieved by varying the parameters values. Thus, variable tolerances are introduced in the recognition proces s. Specifically, tolerances to slight shape variations and intensity Variat ions of the objects (alphabetic characters) are analysed in this work. Rang es for the two control parameters are found in each case in order to achiev e either an increase or a relaxation in the system's discrimination capabil ity. The developed application is extended to colour pattern recognition by multichannel correlation. In this case, four further applications with sel ective discrimination capability are developed: pattern recognition with hi gh discrimination capability for shape variations and some tolerance to col our Variations and, vice versa, pattern recognition with high discriminatio n capability for colour variations and some tolerance to slight shape varia tions; pattern recognition with high discrimination for both shape and colo ur, and, finally, a tolerance to slight variations in both shape and colour . (C) 1999 Elsevier Science B.V. All rights reserved.