Using Surface Metrology to Accurately Match Bullets to Firearms

This article outlines how to establish whether an unknown bullet found at a potential crime scene was fired with a specific firearm. Bullets that are fired with the firearms are engraved with certain distinguishing marks. Therefore, the firearm from a particular event can be linked by firing a second known bullet and then comparing the marks.

Fired bullets contain several land impressions (land-engraved areas) that rely on the barrel rifling, as shown in Figure 1. In order to ensure an exact match, all of the impressions should match both in sequence and form around the two bullets (Figure 2).

Figure 1. Impressions left on a Beretta bullet.

Figure 2. Fired Beretta bullets have six land impressions that are analyzed to find a match in both form and sequence.

The aim is to determine the land-engraved areas and match them up against the two bullets (Table 1). If it is not possible to obtain a complete match, this would probably indicate that the bullets have not been fired from the same firearm.

Table 1. Ideal land impression matches between two bullets, well known (B1) and unknown (U1).

Unkown Bullet 1 (U1) Know Bullet (B1)
B1L1 B1L2 B1L3 B1L4 B1L5 B1L6

Establishing a Match Between Two Land Impressions

A new method has been developed by the Alabama Department of Forensic Sciences (ADFS) to compare land impressions using the Sensofar Forensic Module,  SensoMAP software and the S neox 3D metrology system. The steps involve:

  1. Measuring the entire length of the land impression and an extended topography about 1 mm wide (Figure 3)
  2. Filtering is done to choose a region of interest and acquire separate properties of the surface (Figure 4)
  3. Extracting a profile and directly comparing both bullets (Figure 4)

Figure 3. Land impression and scanned regions on two bullets fired with the same firearm.

Figure 4. SensoMAP data processing on both scanned areas for bullet 1 and 2.

The following steps are then performed:

  • Calculate the cross-correlation function (CCF): CCFMAX is the main parameter and indicates one if both profiles are similar. Figure 4 shows both profiles, where the CCFMAX value is 0.957.
  • Calculate the Signature Difference (DS): DS = R’Q2 / RQ2

NIST recommended this parameter for measuring signature variations. RQ represents the roughness from bullet 1, while R’Q is the roughness of the variation between mean profiles (bullet 1 and 2). DS is evidently zero if the two profiles are exactly the same; however, it increases with increasing difference between the two profiles.

Profiles obtained from the measurements are shown in Figure 4. These measurements were acquired after matching the land impressions of both bullets fired from the same firearm. Two consistent parameters have been developed by NIST to elucidate the extent of similarity. Table 2 shows the CCFMAX and DS values for all six land impressions present on the two known bullets.

Table 2. Similarity parameters for two known bullets

B1L1 B1L2 B1L3 B1L4 B1L5 B1L6
B2L1 B2L2 B2L3 B2L4 B2L5 B2L6
CFFMAX 0.957 0.896 0.902 0.855 0.896 0.949
DS(%) (k) 11.4 17.2 57.5 67.0 26.8 22.8

Figure 4 shows profiles (B1L1 and B2L1), where DS is 11.4% and CCFMAX is 0.957. It should be noted that CCFMAX is high for other two profiles illustrated in the Table 2, while a larger inconsistency is seen in DS.

Comparing Unknown Profiles

The values are used to evaluate a land-engraved area of a known bullet against all the land-engraved areas of an unknown bullet. A contour providing six pairs of profiles is shown in Figure 5, where plotting of the known profile is done towards the left of each unknown profile. It was observed that the known profile related to the B1L1 land impression matches well with the unknown profile related to the U1L1 land impression.

Figure 5. Comparison of a known profile (B1) to a set of unknown profiles (U1).

This conclusion was confirmed when the two parameters DS and CCF for the non-matching and matching land impressions were compared (Table 3). L1 is the land impression where the differences were found to be smaller, while the differences were quite high for the other land impressions.

Table 3. CCF and DS values for the matching and non-matching land impressions

U1L1 U1L2 U1L3 U1L4 U1L5 U1L6
CFF 0.915 0.706 0.752 0.602 0.773 0.727
Difference from CFFMAX 0.042 0.196 0.144 0.253 0.123 0.252
DSU (%) 11.4 130 67.0 346 49.8 107
Difference from DS(k) 0.0 72.5 49.8 279 23 84.2


  • 3D optical comparative measurements help to corroborate casework determinations, which were made using comparison microscopy and other similar standard analysis
  • A complete solution is provided by Sensofar Forensic Module, SensoMAP software, and S neox 3D metrology system Confocal methods can be used to acquire highly accurate 3D measurements, and sophisticated software can be used to automatically measure major analysis parameters such as DS and CCF
  • With the help of 3D confocal microscopy measurements, objectivity can be added to Firearms and Tool Marks analysis

This information has been sourced, reviewed and adapted from materials provided by Sensofar.

For more information on this source, please visit Sensofar.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this article?

Leave your feedback