Page 130 - Automated Fingerprint Identification Systems (AFIS)

FROM PRINT TO IDENTIFICATION 115

in location of minutiae placement, change in number of minutiae placed, area
of the latent print in which minutiae are extracted by coder or manually, or any
combination of the above. These changes are encompassed in case level data
such as geographic area, image data such as minutiae placement, and finger
data such as finger number and pattern. Since capturing and loading the latent
print image into AFIS may have taken a significant amount of time and thus
ideally would not be repeated, the option of relaunching, or performing a
second search, can be very appealing. The variables offer numerous search
options. From a production standpoint, however, more time can be spent in
searching through relaunches than in recapturing the image using photo-
graphic techniques.
If there is no ident after a reasonable number of searches have been com-
pleted, it may be because the subject had never been fingerprinted and
enrolled in the searched database, or perhaps the subject is in the database but
the image quality does not display sufficient minutiae to match against the
minutiae from the latent image. In such cases, the examiner may move the case
to the unsolved latent file, with the anticipation that the subject will be enrolled
for the first time or enrolled with better images and minutiae in the future.
See Table 5.2 for a comparison of pre- and post-AFIS latent print search
processes.

5.4.3 LATENT TO LATENT SEARCHES

The latent examiner may also initiate a latent/latent search. In this search, the
examiner, recognizing that the latent print remains unidentified, looks for a
match with another latent image from the unsolved latent file. If there is a
match with a latent print in this file, the examiner can determine who initiated
the search off the other latent print. By contacting the other inquiring agency,

Table 5.2
Process Pre-AFIS Post-AFIS Latent Print Searches—
Pre- and Post-AFIS
Search time Months Minutes
Suspects or other identifying Required Not required
data
Number of candidates Thousands Less than five candidates per
search (average)
Verification time 2,100 hours Minutes
Number of images compared Thousands Millions
against
Searched by Examiners Computer
Verified by Examiners Examiners