Page 97 - Automated Fingerprint Identification Systems (AFIS)

82 AUTOMATED FINGERPRINT IDENTIFICATION SYSTEMS

search. Finally, it is the latent examiner, following years of extensive training
and experience, who makes the identification.
For latent print searching, the AFIS system eliminates the need to look at
individual tenprint records selected by following the rules of the Henry or
American Classification System. Instead of selecting tenprint candidates based
on the fingerprint classification of the latent print, searching the CCH for a list
of suitable candidates, and calling for and examining individual tenprint cards,
AFIS does the sort. For latent print examiners, the AFIS system provided a
quantum leap in latent print identification. Before AFIS, many latent print
examiners had only suspect or elimination prints to compare against the latent
print found at the crime scene. Larger police agencies may have had a filing
cabinet of tenprint records of known offenders, classified by either the Henry
or the American Classification System, and perhaps further divided by type of
crime and location, i.e., burglary in midtown. If there was no suspect prints to
compare, the latent print examiner would classify the latent print and check
the files that corresponded to that classification. They might also check a file
of other latent print images to see if there was a match and if the same unknown
individual was still committing crimes. Other systems, such as Kodak’s Mira
Code and the Computer Assisted Latent Print System (CALPS), were used, but
they were all of limited effectiveness and very labor intensive.
The following may be an extreme example, but it illustrates the difficulties
of latent print identification before the advent of AFIS technology. A large met-
ropolitan city was threatened by a serial killer, who was randomly shooting, and
sometimes murdering, citizens. A latent print found at one of these murders
was examined, and enough ridge detail was present to identify classification,
pattern, and ridge count. Examiners queried the CCH for a list of matching
records, with some addition and elimination on the number of ridges. A list of
approximately 15,000 candidates was returned. The tenprint cards for all 15,000
were pulled, examined, and returned. The process took months. Today
that same process, searching a much larger database, could be completed in
minutes.
If a latent search results in no candidate or no match, the search parameters
can be changed and the database searched again. For example, if the first
attempt searched only the records of persons charged within a particular
county, a second search, or relaunch, could include records from adjacent
counties, all counties within a region, or within the entire state.
Latent examiners may perform as many individual searches as necessary. A
latent case can have as many as 25 individual lifts, each of which may be
launched (i.e., the minutiae are searched against the minutiae database) three
times before being entered into the unsolved latent file. A search filtered by
parameters such as county, sex, or crime group will search the matchers in times