Page 197 - Automated Fingerprint Identification Systems (AFIS)
P. 197

182  AUTOMATED FINGERPRINT IDENTIFICATION SYSTEMS



                                 The recent push for improved national security has caused the number and
                              type of persons being fingerprinted, both for criminal and civil purposes, to
                              increase. A few years ago firefighters were not printed; now they are. Employ-
                              ees such as child care workers, mental health aides, etc., are now being finger-
                              printed and their records are retained on a database.

                              8.5.2.4 How Many Records Are Added, Deleted, or Updated Annually?
                              How much does the database change every year? New records are created for
                              subjects who were never fingerprinted in the past, additional tenprint records
                              are added for new events such as arrests, and records are deleted. Records may
                              be deleted due to notification of death, court order, age limitation, or being in
                              a category no longer considered appropriate for fingerprint image retention.
                                 The addition of new records presents new opportunities for making an iden-
                              tification that was not possible in the past. If the owner of a latent print had
                              not been enrolled in the database, there was no chance of making the identi-
                              fication. If, however, the subject was arrested after the search has been com-
                              pleted, there are at least two options for re-searching the latent print. The first
                              is the tenprint to unsolved search, in which new records are automatically
                              searched against a database consisting of latent images that were not identified
                              on an AFIS search. These images are maintained on a separate database. When
                              a new record meets a threshold score, a message is sent to the examiner to look
                              at the case. The second option is to recapture, or relaunch, a latent print search
                              at some time after the original search. This option is valuable since there may
                              have been a new addition to the database during that time period that did
                              produce a candidate in the UL/TP file. There may also have been changes to
                              the structure of the database itself, which would allow a more exacting search
                              and match than was previously possible.
                                 The preceding paragraphs have described human and electronic parameters
                              that may affect the latent print identification. Who collects the latent print evi-
                              dence at the crime scene, taking elimination prints, the quality and number of
                              records on the AFIS database are all factors that affect the number of identifi-
                              cations made both with AFIS technology and by other means. The next section
                              describes some of the influences on determining the statistics of reporting
                              identifications.



                              8.5.3 COUNTING LATENT PRINT IDENTIFICATIONS
                              What is considered an ident? For criminal processing purposes, an ident is a
                              latent print image that is matched by standard techniques against a known print
                              image. Counting idents is another matter. Assume an agency collects 100 latent
                              prints from 80 crime scenes for 80 cases. From these 100 latent prints, the latent
   192   193   194   195   196   197   198   199   200   201   202