NCJ Number
57684
Date Published
1978
Length
178 pages
Annotation
TESTING METHODS AND RESULTS ARE REPORTED FOR THE DEVELOPMENTAL MODEL OF HANDWRITING PERSONAL IDENTITY VERIFICATION FOR THE PURPOSE OF ESTABLISHING EFFECTIVE CONTROLLED ACCESS TO AN AREA.
Abstract
AS PART OF ITS EFFORT TO DEVELOP EXTERNAL PHYSICAL SECURITY SYSTEMS FOR THE DEPARTMENT OF DEFENSE, THE UNITED STATES AIR FORCE, UNDER ITS BASE AND INSTALLATION SECURITY SYSTEM (BISS) PROGRAM, ACQUIRED ADVANCED DEVELOPMENT MODELS OF THREE AUTOMATED IDENTITY VERIFICATION SYSTEMS FOR USE IN ENTRY CONTROL. THIS VOLUME REPORTS TEST RESULTS ON ONE OF THOSE SYSTEMS, THE AUTOMATIC HANDWRITING VERIFICATION (AHV) SYSTEM. TO USE THE AHV TECHNIQUE, AN INDIVIDUAL DESIRING ACCESS INTO A CONTROLLED AREA ARRIVES AT THE ENTRY CONTROL POINT AND ENTERS HIS CLAIMED IDENTITY IN THE FORM OF A FOUR DIGIT IDENTIFICATION NUMBER THROUGH A KEYBOARD AT THE ENTRY TERMINAL. THE AUTOMATIC SYSTEM THEN ASKS THE ENTRANT TO PROVE HIS IDENTIFY THROUGH HANDWRITING, BASED UPON PREPROCESSING OF THE PRESSURE VERSUS TIME HISTORY OF A SIGNATURE. THE PURPOSE OF THE TESTING WAS TO DETERMINE THE TYPE I ERROR RATE (FALSE REJECTION OF AUTHORIZED PERSONNEL), RANDOM TYPE II ERROR RATE (FALSE ADMITTANCE OF UNAUTHORIZED PERSONNEL), AND THROUGHPUT. THE TESTS WERE CONDUCTED BOTH AT THE VERIFICATION LABORATORY AT THE MITRE CORPORTATION AND IN THE FIELD AT THE WEAPONS STORAGE AREA AT PEASE AIR FORCE BASE, N.H. FROM TEST RESULTS, IT IS CONCLUDED THAT ABOUT 0.5 PERCENT OF THE POPULATION WOULD REQUIRE SPECIAL TRAINING TO KEEP THEIR TYPE I ERRORS FROM BECOMING A NUISANCE. ALTHOUGH THE TYPE I ERROR RATE WAS HIGHER FOR FEMALES IN THE TESTING, IT IS BELIEVED THAT, BASED SOLELY ON THE AHV ALGORITHM DESIGN, NEITHER TYPE I NOR TYPE II ERROR RATES SHOULD BE AFFECTED BY THE SEX OR HANDEDNESS OF THE USER. CHANGES MADE FOR THE FIELD TEST MADE ITS RESULTS QUESTIONABLE, AND SIGNIFICANT CHANGES ARE REQUIRED TO BRING A SURFACE-EQUPPED AHV SYSTEM TO A POINT WHERE FURTHER SERVICE TESTS ARE APPROPRIATE. THE EVIDENCE DOES NOT SUGGEST, HOWEVER, THAT SUCH A MODIFIED SYSTEM WOULD MEET THE BISS ERROR RATE REQUIREMENTS. THE AHV SYSTEM IS STILL IN AN EARLY STATE OF EVOLUTION. THE APPENDIXES EXPLAIN COMPUTATION PROCEDURES. GRAPHIC AND TABULAR DATA ARE PROVIDED, ALONG WITH A LIST OF REFERENCES USED FOR APPENDIX MATERIAL. (RCB)