BIOMETRIC PERSONAL IDENTIFICATION SYSTEM BASED ON THE IRIS ANALYSIS.

摘要

THE PICTURE ANALYSIS ALGORITHMS FIND THE IRIS IN A LIVE VIDEO IMAGE (10) OF A FACE OF A PERSON, AND CODIFY ITS TEXTURE WITHIN AN "IRIS CODE" (24). THE IRIS TEXTURE IS REMOVED FROM THE MULTIPLE SCALE SCREEN PICTURE BY A SELF-SIMILAR SET OF SQUARE SCREEN FILTERS DEFINED IN A POLAR COORDINATED SYSTEM WITHOUT DIMENSIONS. THE SIGN OF THE PROJECTION OF MANY DIFFERENT PARTS OF THE IRIS ON THESE FILTERS DETERMINES EACH BIT IN AN IRIS CODE. COMPARISONS BETWEEN THE CODES ARE EASILY IMPLEMENTED BY THE O-EXCLUSIVE LOGIC OPERATION (XO). THE MODEL RECOGNITION IS CARRIED OUT BY THE COMBINATION OF SIGNAL PROCESS METHODS WITH THE STATISTICAL DECISION THEORY, LEADING TO A STATISTICAL INDEPENDENCE TEST BASED ON A SIMILARITY METHOD (HAMMING DISTANCE) THAT IS CALCULATED BY CALCULATING X LEVEL OF OBJECTIVE CONFIDENCE (30) ASSOCIATED WITH THE IDENTIFICATION DECISION.