6 FpVTE 2012—Class C

Results in this section involve different impression types with all ten fingers. When plain impression fingerprints were used, the configuration contained left slap, right slap, and left and right thumbs (i.e., finger positions 11–14, or a 4-4-1-1 configuration), as both probes and references. Probes were searched against enrollment databases of 5 000 000 subjects containing all ten fingers.

The datasets in this section are equivalent to those used in NIST FpVTE 2012 (Class C). Detailed information about FpVTE 2012 can be found in NIST IR 8034.

Notes:

  • No examiner extended feature set data was provided with the images.
  • Slap segmentation, if required for the plain impressions, was performed by tech5impl+0005.

6.1 Template Generation

The approximate total number of records that underwent template generation along with a tally of records that failed to process are shown in Table 6.1. Each template was generated by a single function call providing tech5impl+0005 all of the listed image types.

Table 6.1: Summary of template generation for FpVTE 2012—Class C.
Image Contents Template Type Failure to Extract ≈Total
Ten Fingers (Plain) Probe 0 30 000
Ten Fingers (Roll) 0 30 000
Ten Fingers (Plain) Reference 0 5 000 000
Ten Fingers (Roll) 0 5 000 000

6.2 Search

The probe templates from Table 6.1 were searched against two enrollment databases of 5 000 000 subjects containing images as specified in the reference template rows of Table 6.1. Approximately one-third of the probes had a corresponding mate in the enrollment database.

6.2.1 DET

The DET plot in Figure 6.1 shows the tradeoff of errors of tech5impl+0005 when searching probe templates from FpVTE 2012—Class C against enrollment databases of 5 000 000 subjects created from the reference templates from FpVTE 2012—Class C where, for approximately one-third of the probes, a single mated identity was present. Tabular versions of FNIR at select FPIR can be viewed in Table 6.2.

DET when searching probe templates from FpVTE 2012---Class C against a enrollment databases generated from reference templates from FpVTE 2012---Class C.

Figure 6.1: DET when searching probe templates from FpVTE 2012—Class C against a enrollment databases generated from reference templates from FpVTE 2012—Class C.

Table 6.2: FNIR values from the DET plotted in Figure 6.1.
Probe Content Reference Content FPIR  ≤ 0.001 FPIR  ≤ 0.005 FPIR  ≤ 0.01
Ten Fingers (Plain) Ten Fingers (Plain) 0.0018 0.0011 0.0010
Ten Fingers (Plain) Ten Fingers (Roll) 0.0006 0.0004 0.0004
Ten Fingers (Roll) Ten Fingers (Roll) 0.0003 0.0001 0.0001
Table 6.3: Similarity score thresholds from the DET plotted in Figure 6.1.
Probe Content Reference Content FPIR  ≤ 0.001 FPIR  ≤ 0.005 FPIR  ≤ 0.01
Ten Fingers (Plain) Ten Fingers (Plain) 0.2197 0.1867 0.1770
Ten Fingers (Plain) Ten Fingers (Roll) 0.1629 0.1264 0.1202
Ten Fingers (Roll) Ten Fingers (Roll) 0.1617 0.1343 0.1278

6.2.2 CMC

The CMC plot in Figure 6.2 show the FNIR of tech5impl+0005 when searching probe templates from FpVTE 2012—Class C against enrollment databases of 5 000 000 subjects created from the reference templates from FpVTE 2012—Class C where, for approximately one-third of the probes, a single mated identity was present. Tabular versions of FNIR at select ranks can be viewed in Table 6.4.

CMC when searching probe templates from FpVTE 2012---Class C against enrollment databases generated from reference templates from FpVTE 2012---Class C.

Figure 6.2: CMC when searching probe templates from FpVTE 2012—Class C against enrollment databases generated from reference templates from FpVTE 2012—Class C.

Table 6.4: FNIR values from CMC plotted in Figure 6.2.
Probe Content Reference Content Rank 1 Rank  ≤ 2 Rank  ≤ 5 Rank  ≤ 10 Rank  ≤ 50 Rank  ≤ 100
Ten Fingers (Plain) Ten Fingers (Plain) 0.0042 0.0010 0.0005 0.0004 0.0004 0.0004
Ten Fingers (Plain) Ten Fingers (Roll) 0.0037 0.0008 0.0003 0.0001 0.0001 0.0001
Ten Fingers (Roll) Ten Fingers (Roll) 0.0034 0.0003 0.0003 0.0001 0.0001 0.0001