3 Identification Flats (“ThreeInch” Data)

3.1 Segmentation Timing

All algorithms are run over a small fixed corpus of ThreeInch images to estimate the total runtime of the evaluation. To be evaluated under SlapSeg III, algorithms must segment the timing corpus, on average, in under 1 500 milliseconds. This maximum reference time is documented in the SlapSeg III test plan, and is subject to change.

Box plots of segmentation times are separated by hand in Figure 3.1, with tabular representations are enumerated in Table 3.1. Results are reported in milliseconds

Figure 3.1: Box plots of elapsed time in milliseconds when segmenting the ThreeInch timing test corpus, separated by slap orientation.

Table 3.1: Elapsed time in milliseconds when segmenting the ThreeInch timing test corpus, separated by slap orientation.
	Right	Left	Thumbs	Combined
Minimum	33	32	13	13
25%	37	38	18	36
Median	39	39	19	38
75%	43	43	21	42
Maximum	1 419	1 412	449	1 419

3.2 Segmentation Centers and Dimensions

3.2.1 Segmentation Centers

The plots in this section show the distribution of segmentation position centers (x, y) for ThreeInch data. At the top of each figure is a combined plot for all finger positions of a given hand orientation. These figures are isolated in plots faceted at the bottom of the figure.

Plots of segmentation centers for the right hand ThreeInch data are shown in Figure 3.2, for the left hand in Figure 3.3, and for thumbs in Figure 3.4. Blank lines that may appear in the plots are not rendering artifacts. Rather, they are indicative of image downsampling. Centers have been normalized to 500 pixels per inch.

Points in each plot are plotted with a semi-transparent opacity. This results in points of particular color appearing “darker” to indicate a higher frequency of the observed value, while “lighter” points indicate a lower observed frequency.

Figure 3.2: Segmentation centers for right hand ThreeInch data.

Figure 3.3: Segmentation centers for left hand ThreeInch data.

Figure 3.4: Segmentation centers for thumb ThreeInch data.

3.2.2 Segmentation Dimensions

The plots in this section show the distribution of segmentation position widths and heights for ThreeInch data. At the top of each figure is a combined plot for all finger positions of a given hand orientation. These figures are isolated in plots faceted at the bottom of the figure.

Plots of segmentation position dimensions for the right hand ThreeInch data are shown in Figure 3.6, for the left hand in Figure 3.5, and for thumbs in Figure 3.7. Blank lines that may appear in the plots are not rendering artifacts. Rather, they are indicative of image downsampling. Dimensions have been normalized to 500 pixels per inch.

Figure 3.5: Segmentation position dimensions for left hand ThreeInch data.

Figure 3.6: Segmentation position dimensions for right hand ThreeInch data.

Figure 3.7: Segmentation position dimensions for thumb ThreeInch data.

3.3 Detailed Segmentation Statistics

This section shows detailed results of segmentation of ThreeInch data. Values in each table are the percentage that the variable in the left-most column was correctly segmented.

Each table has three columns of percentages. The Standard Scoring column shows the percentage of correctly-segmented positions based on the scoring metrics defined in the SlapSeg III scoring document. The Ignoring Bottom Y column shows how the percentage would change if the threshold for the bottom Y coordinate of the segmentation position was ignored. Similarly, the Ignoring Bottom X and Y columns shows how the percentage would change if only the top, left, and right sides of the segmentation position were considered. These two supplemental columns are included because it has traditionally been difficult to determine the exact location of the distal interphalangeal joint.

Table 3.2 shows how successful sequence+0002 segmented fingers for each subject in the test corpus. Table 3.3 shows success for specific finger positions over the entire test corpus. Similarly, Table 3.4 shows success for segmenting the same finger position from both hands.

The remainder of the tables show success per subject when considering combinations of subsets of the fingers on each slap image. Table 3.5 shows success for combinations of all fingers, Table 3.6 for just the index and middle fingers, and Table 3.7 for all except the little finger.

Table 3.2: For each subject, the percentage that at least *Number of Fingers* fingers were correctly segmented, regardless of hand, for a maximum of eight correctly-segmented fingers. In *Standard Scoring*, scoring rules are followed exactly. In *Ignoring Bottom Y*, the bottom left and bottom right Y coordinates are ignored. *Ignoring Bottom X and Y* only checks the locations of the top left and top right coordinates.
Number of Fingers	Standard Scoring	Ignoring Bottom Y	Ignoring Bottom X and Y
1	99.8	99.8	99.8
2	99.4	99.5	99.6
3	98.2	98.3	98.4
4	97.6	97.9	98.1
5	95.6	95.8	95.9
6	94.9	95.6	95.8
7	93.3	95.1	95.5
8	88.7	93.6	95.0
9	75.6	88.5	92.2
10	50.3	70.3	79.1

Table 3.3: For all subjects, percentage that a particular friction ridge generalized position was correctly segmented. In *Ignoring Bottom Y*, the bottom left and bottom right Y coordinates are ignored. *Ignoring Bottom X and Y* only checks the locations of the top left and top right coordinates.
Finger	Standard Scoring	Ignoring Bottom Y	Ignoring Bottom X and Y
Right
Thumb	82.4	94.9	95.3
Index	97.2	98.2	98.8
Middle	94.8	97.3	99.0
Ring	92.5	95.2	97.8
Little	95.0	96.4	97.4
Left
Thumb	80.1	96.0	96.4
Index	95.8	96.8	97.6
Middle	93.7	96.2	98.7
Ring	91.4	93.3	97.4
Little	95.0	96.1	97.1

Table 3.4: Percentage that a particular type of fingerprint was correctly segmented on *Either* or *Both* hands. In *Ignoring Bottom Y*, the bottom left and bottom right Y coordinates are ignored. *Ignoring Bottom X and Y* only checks the locations of the top left and top right coordinates.
Fingers	Standard Scoring	Ignoring Bottom Y	Ignoring Bottom X and Y
Thumb
Either	90.8	99.1	99.2
Both	71.7	91.9	92.6
Index
Either	99.2	99.5	99.7
Both	91.2	92.9	94.2
Middle
Either	98.4	99.2	99.7
Both	87.7	91.7	95.4
Ring
Either	97.7	98.5	99.5
Both	83.8	87.5	93.1
Little
Either	99.0	99.3	99.5
Both	88.4	90.6	92.3

Table 3.5: Percentage of segmentation success by hand for combinations of all ten fingers of a ThreeInch slap. In *Ignoring Bottom Y*, the bottom left and bottom right Y coordinates are ignored. *Ignoring Bottom X and Y* only checks the locations of the top left and top right coordinates.
Fingers	Standard Scoring	Ignoring Bottom Y	Ignoring Bottom X and Y
Right
Any	99.6	99.7	99.7
At Least Two	98.2	98.4	98.4
At Least Three	97.0	98.0	98.2
At Least Four	92.1	95.9	97.1
All Five	66.6	81.1	86.0
Left
Any	99.6	99.7	99.8
At Least Two	98.1	98.3	98.3
At Least Three	96.6	97.7	98.0
At Least Four	90.6	94.8	96.6
All Five	62.8	79.0	85.4

Table 3.6: Percentage of segmentation success by hand when only considering combinations of index and middle fingers. In *Ignoring Bottom Y*, the bottom left and bottom right Y coordinates are gnored. *Ignoring Bottom X and Y* only checks the locations of the top left and top right coordinates.
Fingers	Standard Scoring	Ignoring Bottom Y	Ignoring Bottom X and Y
Right
Either	99.2	99.6	99.7
Both	92.7	95.8	98.1
Left
Either	98.9	99.3	99.6
Both	90.6	93.8	96.7

Table 3.7: Percentage of segmentation success by hand when only considering combinations of index, middle, and ring fingers. In *Ignoring Bottom Y*, the bottom left and right Y coordinates are ignored. *Ignoring Bottom X and Y* only checks the locations of the top left and right coordinates.
Fingers	Standard Scoring	Ignoring Bottom Y	Ignoring Bottom X and Y
Right
Any	99.6	99.8	99.8
At Least Two	97.6	98.8	99.5
All Three	87.2	92.2	96.4
Left
Any	99.4	99.6	99.7
At Least Two	96.7	97.9	99.1
All Three	84.8	88.8	94.8

3.4 Handling Troublesome Images

3.4.1 Capture Failures

Segmentation algorithms may refuse to process an image. This may happen for a technical reason (e.g., the algorithm cannot parse the image data), or for a practical reason (e.g., the hand in the image is placed incorrectly). These failure scenarios are the result of capturing improper image data. In these types of scenarios, it is important to examine the cause of the failure. With many live scan capture setups, segmentation is performed immediately after capture. If an algorithm can detect that it won’t be able to segment an image due to a technical or practical issue, it can alert the operator to perform a recapture before the subject leaves.

The SlapSeg III API encourages algorithms to identify these failure reasons by specifying pre-defined deficiencies in the image. Algorithms should attempt segmentation even if an image deficiency is encountered if at all possible. Note that SlapSeg III guarantees well-formed image data, so failures to parse are not an indicator of the data provided.

Reasons for capture-type failures reported by sequence+0002 are enumerated in Table 3.8.

Table 3.8: Count of self-reported capture-type failure reasoning.
Failure Reason	Images
Request Recapture (Attempt)	119

In situations where the algorithm feels that the presented image should be recaptured (Table 3.8), one or more image deficiencies must be identified. These deficiencies are enumerated in Table 3.9. At this point, NIST does not have a groundtruth of image deficiencies, but plans to update this table with the accuracy of deficiency observations in the future.

Table 3.9: Count of image deficiencies reported when requesting a recapture.
Deficiency	Count
Image Quality	119

3.4.1.1 Recovery

When encountering a segmentation failure, SlapSeg III algorithms are encouraged to provide a best-effort segmentation when possible. In some cases, that best-effort may be correct, which reduces the amount of images that need to be manually adjudicated by an operator.The result of such best-effort segmentations are shown in Table 3.10.

Out of 119 recovery attemptssequence+0002 attempted 327 segmentations of fingers and skipped 145 fingers. More information about skipped fingers can be found in Table 3.11.

Table 3.10: Results of best-effort segmentation when `sequence+0002` reported segmentation failure (327 best-effort attempts).
Standard	Ignoring Bottom Y	Ignoring Bottom X and Y
55.7	58.4	61.8

3.4.2 Segmentation Failures

Even if an algorithm accepts an image for processing, it can still fail to process one or more fingers from the image, regardless of if the algorithm requested a recapture and provided best-effort segmentation.

The SlapSeg III API allows algorithms to communicate reasons for failure to process these fingers. In some cases, the distal phalanx in question might not be present in the image due to amputation or being placed outside the platen’s capture area. It is imperative that the segmentation algorithm correctly report this as failing to segment the correct friction ridge generalized position without disrupting the sequence of valid positions present in the image. This can help prompt an operator to recapture or record additional information about the subject.

In SlapSeg III, a number of images are missing fingers or otherwise have fingers that will not be able to be segmented. Reasons for segmentation failures reported by sequence+0002 are enumerated in Table 3.11.

Table 3.11: Count of self-reported segmentation failure reasoning.
Failure Reason	Fingers
Finger Found, but Can’t Segment	145
Finger Not Found	0
Vendor Defined	0

3.4.3 Identifying Missing Fingers

A small portion of the test corpus in SlapSeg III are missing fingers. Table 3.12 shows how successful sequence+0002 was in correctly determining if a finger was missing. The Missed row shows when a segmentation position was returned for a missing finger. All possible failure reasons are enumerated, but are not considered Correctly Identified because the algorithm specified failure for a reason other than the finger not being found.

Table 3.12: Performance of `sequence+0002` at detecting fingers missing from an image.
Result	Percentage
Missed	65.0
Correctly Identified	0.0
Other Failure: Finger Found, but Can’t Segment	35.0
Other Failure: Vendor Defined	0.0
Other Failure: Segmentation Not Attempted	0.0

3.4.4 Sequence Error

Sequence error occurs when a fingerprint is segmented from an image but assigned an incorrect finger position (e.g., segmenting a right middle finger but labeling it a right index finger). Table 3.13 shows cases in which a segmentation position was returned that matched a ground truth segmentation position for a different finger in the same image.

Table 3.13: Percentage of images in the dataset where one or more segmentation positions correctly matched an incorrect finger position within the same image, indicating sequence error.
Hand	Standard Scoring	Ignoring Bottom Y	Ignoring Bottom X and Y
Left	0.06	0.06	0.07
Right	0.07	0.07	0.07
Thumbs	0.03	0.03	0.03
Combined	0.06	0.06	0.06

3.5 Determining Orientation

An optional portion of the SlapSeg III API asked participants to determine the hand orientation of an image. Participants were provided the kind (e.g., Identification Flat) and needed to determine whether the image was of the left hand, right hand, or thumbs.

Overall Three Inch accuracy: 97.7%

Table 3.14: Percentage of accuracy when determining hand orientation of a three inch image. The first column indicates the true hand orientation. Subsequent columns indicate the percentage of the time in which the indicated hand orientation was hypothesized.
	Left	Right	Thumbs
Left	99.3	0.6	0.1
Right	0.8	99.2	0
Thumbs	3	2.5	94.5