Page 50 - Automated Fingerprint Identification Systems (AFIS)

HISTOR Y OF AUTOMATED FINGERPRINT IDENTIFICATION SYSTEMS 35

agencies could not only search the FBI, but also obtain a criminal history record
(rap sheet) from another state. Criminal identification information was begin-
ning to flow more freely. As NCIC and other identification systems grew, so did
the realization that standardization would be mandatory if the potential of
information exchange was to be realized.
The introduction of large mainframe computers in the 1960s and 1970s
marked a milestone to this end, as data could be kept in a storage system that
did not depend on paper and that could be electronically sorted. While these
computers started life as frail equipment subject to breakdown and high main-
tenance, they soon began to prove their worth, particularly in the countless ways
in which data could be parsed.
By this time, many large identification bureaus, including some states, had
a semi-automated identification system in place. The automated portion of the
system was known as the Computerized Criminal History file, or the CCH. The
CCH contained previously submitted (i.e., enrollment) information. For crim-
inal applications, the information might include the subject’s name, date of
arrest, and arresting agency. The record would also include the numerical
classification of each finger image, as defined by either the Henry or the
American Classification System. Images were not electronically stored, but
remained available on the filed tenprint card.
When the identification bureau received a criminal inquiry, the name search
was generally the first search performed. That is, the name on the tenprint card
sent by the inquiring agency would be entered into the mainframe computer.
If a matching name was found in the CCH, the classification of the ten fingers
matching that name would also be printed. (See Chapter 5 for more details on
name- and image-based searches.)
While this was being done, another fingerprint classifier would enter the fin-
gerprint patterns from the submitted tenprint card and classify the ten images
using either the Henry or the American Classification Systems. A clerk would
pull the filed fingerprint cards for the matching names found by the computer
and compare the images on the card with the submitted tenprint record to
determine if there was a match. If so, the CCH was updated and a criminal
history was sent to the inquiring agency. If there was no match based on the
name search, then a technical search of the database was performed. The pat-
terns of all ten fingers were entered, and the computer would produce another
list of candidates whose patterns matched those of the inquiry prints. If no
match was found, a new record would be created.
These systems relied on the classification of finger patterns by classification
experts. Learning the classification patterns required extensive training and
guidance until successfully mastered. While an experienced examiner could