CASE REPORT
Richard W. Vorder Bruegge,1 Ph.D.

Photographic Identification of Denim Trousers
from Bank Surveillance Film1

REFERENCE: Vorder Bruegge RW. Photographic identification
of denim trousers from bank surveillance film. J Forensic Sci
1999;44(3):613–622.
ABSTRACT: The ability to identify denim trousers (e.g., “blue
jeans”) from bank surveillance film through side-by-side comparison
is documented herein. Individual identifying characteristics that are
generated in the manufacturing process and during normal wear-andtear may be recognized on denim trousers and in photographs depicting those trousers. These characteristics may include folds,
creases, and puckering which manifest themselves as high and low
areas (“ridges and valleys”) along and adjacent to the seams and
hems. As the jeans are worn and washed over time, the visibility of
these ridges and valleys is amplified through abrasion of the ridges,
resulting in a loss of dark dye and marked tonal brightening. Given
sufficient abrasion, even small ridges may exhibit sufficient contrast
against the dye-rich background to be recorded on bank surveillance
films, permitting a comparison with trousers recovered from suspects. The application of this technique in a case involving a series of
bombings and bank robberies in Spokane, Washington, is discussed.
KEYWORDS: forensic science, photographic comparison, identification from photographs, clothing identification, identification of
denim clothing, clothing wear marks, bank surveillance film, blue
jeans, side-by-side comparisons, barcode patterns

Whenever an item recovered from a suspect or a suspect’s home
can be associated with the commission of a crime, it can offer
strong circumstantial evidence in the prosecution of the case against
that individual. The association of articles of clothing worn by bank
robbers depicted in bank surveillance images with clothing recovered from suspects is one example of such circumstantial evidence.
The FBI Laboratory’s Special Photographic Unit conducts
forensic examinations of imaging evidence of every kind, including film negatives, photographic prints, video tape recordings, and
digital images. A primary area of examinations includes the photographic comparison of clothing worn by bank robbers depicted in
surveillance images (film or video) with clothing recovered from
1
FBI Laboratory, Investigative Operations Support Section, Special Photographic Unit, Room 3449, 935 Pennsylvania Avenue, N.W., Washington, DC
20535.
* Presented at the 50th Annual Meeting, San Francisco, CA, Feb. 1998.
American Academy of Forensic Sciences.
* This is publication number 98-08 of the Laboratory Division of the Federal
Bureau of Investigation. Names of commercial manufacturers are provided for
identification only, and inclusion does not imply endorsement by the Federal
Bureau of Investigation.
Received 12 Jan. 1998; and in revised form 4 Aug. 1998; accepted 21 Sept.
1998.
Copyright by ASTM Int'l (all rights reserved); Thu Mar 8 09:01:09 EST 2018

suspects. Such examinations have been conducted for approximately 30 years and currently represent more than 25% of the
unit’s caseload. Other major areas of examinations include facial
comparisons, photogrammetric examinations (e.g., bank robber
height determinations), and image analysis through analog and digital image processing techniques.
In 1996, investigation into a series of bombings and bank robberies in the Spokane, Washington area resulted in the identification of four suspects and the recovery of numerous articles of clothing from their homes and vehicles. The clothing, including 27 pairs
of denim trousers (“blue jeans”), was subsequently submitted to the
Special Photographic Unit for comparison with surveillance images from the bank robberies. What follows is a description of the
process through which one of the seized pairs of blue jeans was
identified as the pair worn by one of the bank robbers.
Photographic Comparisons and the Principle of
Individualization
The principle behind photographic comparisons is the same as
that behind the comparison of fingerprints, footwear impressions,
bullets, firearms, or any other kind of physical evidence—the principle of individualization. As described by Tuthill:
“The individualization of an impression [or other piece of
physical evidence] is established by finding agreement of
corresponding individual characteristics of such number and
significance to preclude the possibility (or probability) of
their having occurred by mere coincidence, and establishing
that there are no differences that cannot be accounted
for (1).”
The key element in conducting a successful comparison in any
examination of physical evidence is the ability to resolve the individual characteristics that differentiate the item from all other similar items. This holds for photographic examinations involving
clothing, faces, firearms, vehicles, or other items. When conducting an examination through photographic comparison, two interrelated factors have the greatest impact on the ability to conduct a
successful examination: (1) the quality of the images recorded on
the film (or video tape or digital imaging system); and (2) the presence of individual characteristics on the known object that are sufficiently visible to be identifiable on the recorded images. In the
case of the bank robber’s blue jeans considered herein, both of
these factors were favorable to an individualization.

Downloaded/printed by
Ryan Gabrielson (ProPublica) pursuant to License Agreement. No further reproductions
613authorized.

Copyright © 1999 by ASTM International

 614

JOURNAL OF FORENSIC SCIENCES

Another factor that is important to consider when applying the
principle of individualization to photographic comparisons relates
to apparent differences that can result from the photographic process. This is a factor that most individuals do not actively consider
when examining photographs, video, or other images. When a photograph is taken, a three-dimensional scene is converted into a flat,
two-dimensional representation. When looking at such a photograph, the viewer does not generally interpret the scene depicted as
being two-dimensional, but automatically understands that the
scene is three-dimensional. The perspective depicted in the photograph, including the relative size of known or familiar objects in the
scene, is used by the viewer to interpret the geometric relationships
between objects within the scene and between the camera and the
scene. Furthermore, most viewers understand that the photographic
process is an imperfect one, in which the actual colors and shades
of a scene are rarely replicated perfectly. The meaningful comparison of objects depicted in photographs with items of evidence
must include specific consideration of such changes introduced by
the photographic process.
A final critical factor that must be considered in photographic
comparisons involves the potential for an object to exhibit temporary or permanent changes in its characteristics such as shape or
tonality between the time it was photographed and the time it is examined. Clothing is particularly prone to changes in shape while
being worn, as the material is stretched, folded, or bunched. Similarly, clothing tends to fade after repeated washings, so that differences in the shape or tonality of an object depicted in a photograph
and an object seized as evidence do not necessarily indicate that
they are not the same object. The potential for such differences is
considered by the examiner to such an extent that the most critical
aspect of a photographic comparison frequently involves establishing that observed differences can be accounted for.
Methodology of Clothing Comparisons

(recognizable by the presence of well-defined vertical ribs and
wells), then any submitted denim trousers can be eliminated. However, if an elimination is not immediately forthcoming, then photographic enlargements of the best frames are produced at exposure
and contrast levels that lead to the highest possible detail in the area
of interest. The search for individual identifying characteristics and
the point-by-point comparison with the submitted trousers can then
proceed.
Identifying Characteristics of Denim Trousers
In order to conduct clothing comparisons, examiners in the Special Photographic Unit must be knowledgeable regarding the
means by which clothing develops class and unique characteristics.
This knowledge is obtained, in part, through on-site inspection at
manufacturing facilities throughout the United States, as well as
through technical discussions with manufacturers. Descriptions of
the procedures included herein are based on inspections and discussions in which the author has participated since 1995.
The most general of class characteristics exhibited by denim
trousers include the manufacturer, style, and size. The style and
size may affect additional class characteristics of the trousers such
as cut of the leg (i.e., straight, tapered, boot-cut, etc.), presence and
type of buttons or zippers in the fly (and elsewhere), presence of
rivets, pocket stitching, presence and style of patches or labels on
the waist, and number and type of belt loops.
Unique or individual identifying characteristics of denim
trousers typically arise as defects generated by the manufacturing
process (e.g., drop-stitches—locations where a stitch is missed)
and through normal wear-and-tear. The latter may include paint,
ink, blood, dye or other stains, rips, tears, holes, fold creases, wrinkles, and other wear marks. Such characteristics have been used in
clothing identifications by the Special Photographic Unit for approximately thirty years (2) and were discussed by FBI Special
Agent Joseph M. Avignone in a presentation to the 1989 American
Academy of Forensic Sciences Annual Meeting (“Clothing Identification Analysis of Bank Robbery Films”). Finally, in some cases
individualizing patterns of wear marks can result from creases or
puckers introduced during the manufacturing process.
It is important to note that in making clothing comparisons the
examiner is concerned with the morphology of the individual identifying characteristics and their approximate location relative to
fixed landmarks on the article of clothing. To an extent, the examination is one of pattern recognition. Specific measurements detailing the precise location of identifying marks on clothing are not
necessary for an association to be made, nor can they be made in
the majority of cases due to several factors, not the least of which
is temporal variations in the shape of the clothing such as folding,
billowing, or “bunching up” of the material. Furthermore, photogrammetric analysis of the type used in examining surveillance
images is adequate when measuring linear distances such as
heights, but is less accurate when measuring curved objects such as
a draped trouser leg, whose three-dimensional shape cannot be
uniquely determined from a single image, particularly when the exact location and orientation of that object in space is not known.
This is in contrast to comparisons of impressions in which the impression is typically left on a planar surface or on an object whose
shape can be uniquely defined so that later corrections can be made
for distortions.

In clothing comparison cases, the bank film depicting the unknown bank robber is labeled the “Questioned” item of evidence.
The recovered clothing is the “Known” item since its source is
known. Upon receipt of the bank film and recovered clothing, every frame of the film is examined to locate those frames which best
depict the bank robber and his clothes.
Examination of the original film is critical, since it will contain
the highest level of detail in both the shadows and highlights. Photographic prints produced from bank film through automated processors are generally exposed to produce adequate contrast and image quality for the overall scene, rather than for the specific
portions that contain the bank robber and his clothing. In such machine-processed prints, fine details needed to individualize the
clothing may not be visible due to under- or over-exposure in the
portion of the print depicting the clothing. In an examination using
the original film, various exposures may be tested to maximize the
detail visible in the final print.
It is also important to select frames in which the questioned individual or item is at rest. Motion blur can be a major factor in causing fine detail to be absent from an image. Likewise, if the resolution of an image is less than that necessary to distinguish the fine
detail, then it is impossible to make an identification or elimination
based on that detail.
It is during the review of the film that analysis begins, with the
observation of the class characteristics of the clothing. Once such
Folds and “Barcode” Wear Patterns on Denim Trousers
characteristics are noted on the clothing worn by the bank robber,
preliminary
comparisons
are(allconducted
withThu
theMar
known
items
The side seams, inseams, and hems of denim trousers typically
Copyright
by ASTM Int'l
rights reserved);
8 09:01:09
ESTof2018
clothing.Downloaded/printed
For example, if by
the bank robber is wearing corduroy pants
develop a pattern of puckers or “ridges-and-valleys” which, after
Ryan Gabrielson (ProPublica) pursuant to License Agreement. No further reproductions authorized.

 VORDER BRUEGGE • IDENTIFICATION OF DENIM TROUSERS FROM FILM

615

dyed denim trousers might also be likened to a “railroad track” pattern. However, such a characterization implies a regular spacing
and constant width to the bright and dark patches along the seam or
hem. Inspection of the seam in Fig. 1 reveals that this is not the
case. Instead, the width and spacing of the bright and dark patches
are irregular, as in a barcode. This irregularity is what makes the
barcode pattern useful as an identifying characteristic. Although a
validation study has yet to be performed to test the theory that all
denim trouser barcode seam patterns are unique, it has been observed in numerous examinations that it is possible to distinguish
pairs of jeans from one another based solely on differences in the
patterns along the seams.
In some cases, individual folds or creases are also generated during manufacture or through later processing of the trousers. For example, folding of the trousers and subsequent ironing can lead to a
marked raised crease which is then subject to increased abrasion.
As the abrasion increases, the visibility of the crease against the
background is amplified as the dye-bearing outer layer is removed,
revealing the un-dyed white cotton cores of the yarn underneath.
Alternately, in some cases puckering along the seams and hems can
result in individual low areas which remain dye-rich (and therefore
dark) while the surrounding areas are worn in a uniform fashion,
leaving a single dark feature against a lighter background. Such individual characteristics may be more significant than the barcode
pattern in a photographic comparison since they typically stand out
from the solid background of the denim and are more easily identifiable as random features. This was the case in the analysis of the
Spokane bank robbery film and examples of such characteristics
are discussed below.
Effects of Stitching and “Cross-over” Wear Patterns
FIG. 1—Pucker or “ridge-and-valley” pattern along seam of blue jeans
resembles a “barcode.”

being worn down, will appear as an alternating sequence of light
and dark bands, resembling a “barcode” (Fig. 1). These puckers are
generated during the manufacturing process as the operator pushes
the material being stitched through the sewing machine. This process is not automated and causes material to alternately bunch up
and stretch as it is guided through the sewing machine, creating an
irregular waveform pattern that becomes ingrained in the seam or
hem. Later shrinkage may increase the amplitude of this waveform,
and, therefore, increase the prominence of this pattern.
In contrast to an actual UPC bar code in which the alternating
pattern is pre-defined, no attempt is made by the sewing machine
operator to produce a specific pattern, nor to replicate this pattern
of puckers in subsequent seams or hems. Nor is it likely that such
patterns could be repeated since the operation is performed by
hand. Therefore, the waveform is assumed to represent a unique
manufactured characteristic. Over time, as the trousers are worn
and washed, the raised portions of the waveform become abraded,
which leads to a loss of the dyed surface layer and exposure of the
un-dyed core of the white cotton yarn. When the trousers in question are blue jeans (or any other dark denim trouser, such as black
jeans) the removal of the dark dye and exposure of the white yarn
cores on the ridges leads to a visible brightening of these ridges.
Since the fabric in the valleys retains a dark dyed surface layer between bright, abraded ridges, a recognizable alternating bright and
dark pattern results, which is referred to here as a “barcode” pattern.
It has been
suggested
that reserved);
the alternating
ofEST
bright
Copyright
by ASTM
Int'l (all rights
Thu Mar pattern
8 09:01:09
2018and
dark
areas along the
Downloaded/printed
by seams and hems of blue jeans and other dark-

When two pieces of fabric are brought together to make a seam,
they may be stitched together in one of two ways: they may be
“felled” or they may be “chainstitched.” Many denim trousers are
constructed with felled inseams and chainstitched side seams. A
felled inseam consists of two pieces of fabric that are folded over
upon one another prior to being sewn with a double-stitch. This
leaves a seam that is 4-plies thick, with stitching visible on the outside of the garment. The presence of an extra 3-plies-worth of material along a felled seam plays an important role in the generation
of the barcode pattern by making the seam stand up higher than the
surrounding material, thus exposing it to more wear through abrasion and an associated loss of the dye-rich outer layer.
In a chainstitched side seam, the two pieces of fabric are pinched
together and sewn in such a way that the stitching is not visible on
the exterior of the garment. However, such a chainstitched seam is
only 2-plies thick, and will be weaker than one that is felled. Furthermore, the chainstitch leaves flaps of excess material on the interior of the garment which are free to fold over and lay flat against
the garment. Frequently, the seam is pressed open or “seambusted”
so that one flap lies against the material on each side of the seam.
Alternately, the flaps are sometimes sewn together, creating a single 2-ply flap that is then free to fold down to one side or the other
of the seam. In either case, the extra layers will raise the fabric in
the vicinity of the seam, making these areas more prone to abrasion
than surrounding ones.
In addition to the key role the flaps of excess material play in amplifying the visibility of the barcode pattern along a side seam,
these flaps are also crucial in the generation of another characteristic known as a “cross-over.” The flap or flaps of material do not
necessarily fold to one side or the other along the entire length of
the seam. In some places the flap will fold back to the other side of

Ryan Gabrielson (ProPublica) pursuant to License Agreement. No further reproductions authorized.

 616

JOURNAL OF FORENSIC SCIENCES

the seam, and in the process create an unusually thick portion of
material where the flap effectively stands on end. This point thus
represents the location on the seam with the thickest cross-section
of material, and therefore the most likely point of abrasion and
wear. Furthermore, by changing the side on which the flap lies, the
raised part of the seam that amplifies the barcode pattern will also
change sides or “cross-over.” Exterior and interior views of a
chainstitched side seam that exhibits a cross-over are presented in
Fig. 2.
Cross-over features are observed on denim trousers at locations
along the seam ranging from the hip to just above the hem, with no
apparent correlation between the location of this feature and the
trouser manufacturer or style. In some cases, denim trousers have

been found to exhibit more than one cross-over along a single
seam. Furthermore, it has been observed that the location and occurrence of cross-over features can differ on pairs of denim trousers
owned and worn by the same individual. From these limited observations one might propose that the location and frequency of crossover features are independent of the manufacturer and wearer.
However, a thorough validation study to test this proposal has yet
to be performed. Nevertheless, as with differences in the barcode
patterns, the cross-overs create observable wear patterns that persist over time, making it possible to use differences in the location
and nature of cross-over features to distinguish one pair of denim
trousers from another. Thus, one can use cross-overs as an identifying characteristic in photographic comparisons.

FIG. 2—Cross-over pattern. (a) Exterior view. (b) Interior view—actual view has been reversed to match the orientation of the exterior view shown in
(a).
Copyright by ASTM Int'l (all rights reserved); Thu Mar 8 09:01:09 EST 2018
Downloaded/printed by
Ryan Gabrielson (ProPublica) pursuant to License Agreement. No further reproductions authorized.

 VORDER BRUEGGE • IDENTIFICATION OF DENIM TROUSERS FROM FILM

Spokane Bank Robbery and the Observed Characteristics of the
Robber’s Trousers
On April 1, 1996, two bank robbers entered a branch of the U.S.
Bank in Spokane, Washington. The surveillance cameras were triggered by the bank personnel and several hundred frames depicting
the movements of the robbers in the bank were taken. Many of the
frames depict one robber carrying a large firearm, and wearing a
dark mask, parka, athletic footwear, and denim trousers. Figure 3
contains two of the best views of this robber and his trousers. To
the great benefit of this examination, a number of the frames contained on the bank film depict the robber standing perfectly still
while presenting direct views of his right outseam and left inseam.
Subsequent investigation of this robbery, a second robbery of the
same bank on July 12, 1996, and several related bombings, led to
the identification of four suspects. During searches of their homes,
numerous articles of clothing were recovered, including 27 pairs of
denim trousers, all of which were blue jeans. These articles of
clothing were submitted to the Special Photographic Unit for comparison with the bank film from both robberies.
The bank robber’s parka extends below the waist, effectively
masking the manufacturing characteristics of the belt loops, pock-

617

ets, and fly. However, individual characteristics are recognized on
and near the left inseam, right outseam, and hems (Fig. 4). Shadowing and the prominence of the film grain prevent one from
clearly resolving the left inseam’s barcode pattern (Fig. 4a).
Nevertheless, a number of distinctive individual characteristics
can still be recognized near the left inseam and hem. The first is a
bright linear feature extending up from just left of the inseam-hem
intersection. This narrow linear feature parallels the inseam for a
short distance then angles away from it, well up the calf (“1”). A
second feature consists of a connected set of dark patches in the
general shape of an “H” which span the inseam-hem intersection
(“2”). The right leg of the “H” is relatively broad and is defined on
the right by a bright lineament. The left leg of the “H” is narrower
and abuts the left edge of the inseam, with its outer edge defined by
the linear feature (“1”) described above. A third feature immediately to the left of the “H” consists of a pair of bright, short, lineaments extending from the hem. This pair of marks form a “V”shape that tilts to the left so that the right side is vertical and the left
side extends at an angle of approximately 45-degrees (“3”).
Finally, a pair of bright features resembling the Greek letter “II”
(“4”) are observed to the immediate right of the dark “H” along the
hem.

FIG. 3—(a) Bank surveillance photograph depicting view of Spokane bank robber’s left leg. (b) Bank surveillance photograph depicting view of Spokane
bank
robber’s
right leg.
Copyright
by ASTM
Int'l (all rights reserved); Thu Mar 8 09:01:09 EST 2018
Downloaded/printed by
Ryan Gabrielson (ProPublica) pursuant to License Agreement. No further reproductions authorized.

 618

JOURNAL OF FORENSIC SCIENCES

B

A

FIG. 4—(a) Enlarged portion of Fig. 3(a) showing details of left leg. (b) Enlarged portion of Fig. 3b showing details of right leg. See text for discussion
of features identified by numbers.

On the right outseam a distinctive barcode pattern is observed
extending up from the hem to above the knee (Fig. 4b). This barcode occurs to the left of the crease that defines the edge of the
seam. Above the knee, a “cross-over” is observed (“5”). This crossover represents a location in which the barcode pattern of bright
and dark patches switches to the right side of the crease. The point
where the side seam meets the hem is marked by a broad bright
patch, whose upper right edge is sharply defined along a diagonal
from the side seam to the hem (“6”). From the point where the right
side seam meets the hem up to the cross-over, twenty-one (21) individual bright patches (of various widths and distinctiveness) are
observed (“8–28”). Additional bright patches were noted both
above the cross-over and along the hem to the right of the “H” (“7,
29–30”).
Identification of Submitted Blue Jeans
Having noted the extensive set of individual characteristics on
the blue jeans worn by the parka-clad bank robber in the April 1
bank film, it was then a straightforward process of comparing these

observations with the blue jeans submitted to the Laboratory. The
27 pairs of blue jeans were inspected individually, and the presence
or absence of identifying characteristics similar to those on the
bank robber’s jeans noted. Of these pairs of blue jeans, 26 pairs
were found to have individual characteristics that were sufficiently
different from those exhibited on the bank robber’s trousers to permit their elimination as those worn by the bank robber. The 27 pair,
a pair of JC Penney “Plain Pocket” blue jeans was found to have
characteristics matching all of the individual characteristics noted
on the bank robber’s jeans. Through consultation with the manufacturer it was determined that these jeans had been constructed in
Clarksville, Tennessee in 1991, using standard hand-guided
sewing practices common throughout the blue jean industry. This
consultation confirmed that the observed characteristics could be
expected to be the result of random processes. Therefore, since the
morphology and relative location of these random characteristics
was consistent between the bank robber’s trousers and the submitted blue jeans, the submitted blue jeans were individualized as the
ones worn by the bank robber.
Figure 5 depicts modeling of the blue jeans identified as those

Copyright by ASTM Int'l (all rights reserved); Thu Mar 8 09:01:09 EST 2018
Downloaded/printed by
Ryan Gabrielson (ProPublica) pursuant to License Agreement. No further reproductions authorized.

 VORDER BRUEGGE • IDENTIFICATION OF DENIM TROUSERS FROM FILM

619

FIG. 5—Modeling of JC Penney “Plain Pocket” blue jeans, size 34 (W) ⫻ 30 (L), recovered from suspect’s home. (a) Left inseam. (b) Right side seam.

worn by the bank robber. Figures 6 and 7 show side-by-side, pointby-point comparisons of the blue jeans depicted in the bank surveillance images (left) and the modeled JC Penney “Plain Pockets”
jeans (right).
In a photographic comparison, the modeling photographs serve
an additional purpose beyond simple documentation. They also
serve to confirm that the photographic process is not introducing
any unexpected anomalies into the appearance of the item under
examination. When modeling an article of clothing, the geometry
between the camera and the subject is arranged so as to approximate that between the bank surveillance camera and the questioned
individual. Furthermore, film similar to the original bank surveillance film is used, along with a lens that has a focal length comparable to the one used in the bank. The lighting conditions are also
adjusted to approximate those in the bank at the time of the crime.
If lighting becomes a critical issue in the photographic comparison
of an item, it is sometimes necessary to return to the scene of the
crime and model the item in place at the same time of day and time
of year as in the original image.
In this case, the selected geometry, film, and studio lighting were
sufficient to recreate the resolution, perspective, and lighting conditions in the questioned images and to confirm that no anomalies
were introduced by the photographic process, as well as to document the findings. This permits the examiner to make a side-byside comparison approaching a one-to-one comparison as shown in
Figs.
6 and 7. Recall, however, that an exact one-to-one compariCopyright by ASTM Int'l (all rights reserved); Thu Mar 8 09:01:09 EST 2018

son of a garment through this technique is not usually possible due
to the pliable nature of clothing and the difficulty of replicating every fold and crease in a garment that is being worn by a subject in
motion.
Defense Challenges
During the trials of the suspects, the defense offered an exporter
of used blue jeans as an expert witness to testify that the most
prominent of the characteristics noted above were common to
blue jeans, including the linear wear mark (“1”), the “V” and
“H” marks (“2” and “3”), and the cross-over (“5”). As proof, the defense offered 34 pairs of blue jeans described as having such characteristics. These jeans had been culled from over 400 pairs examined by the exporter at the request of the defense. The exporter
identified the 34 pairs of blue jeans as the ones which best showed
characteristics similar to those of the blue jeans worn by the bank
robber.
For the rebuttal portion of the trial the 34 pairs of blue jeans were
examined by the author. None of the 34 pairs of trousers offered by
the defense shared more than one or two of the prominent characteristics described above in even a gross manner. For example,
none of the pants that had a feature which could be referred to as an
“H” also exhibited features that would correspond to the “V” or
cross-over features. This finding helps emphasize the significance
of being able to identify multiple individual characteristics when

Downloaded/printed by
Ryan Gabrielson (ProPublica) pursuant to License Agreement. No further reproductions authorized.

 620

JOURNAL OF FORENSIC SCIENCES

FIG. 6—Side-by-side comparison of Spokane bank robber’s left leg (“Questioned”) and left leg of model wearing blue jeans recovered from suspect’s
home (“Known”). Numbers refer to features described in text.

attempting to make an association or elimination in such a
comparison.
Furthermore, detailed inspection revealed that most of the characteristics identified by the defense expert as “similar” on these
jeans could be readily distinguished from those on the bank
robber’s jeans. For example, the left upright of the “H” feature on
one pair of proffered jeans was markedly wider than the one observed on the bank robber’s jeans, while on a second pair the right
upright on the “H” feature was not sharply defined by a bright lineament as on the bank robber’s jeans. The ability to make such distinctions in this case further highlights the high quality of the
Spokane film images.
While a rigorous validation study would be necessary to confirm
it, this demonstration did serve to strengthen the proposition that
the individual wear marks observed on the known blue jeans are the
result of random processes and can be considered unique, individualizing characteristics. Furthermore, when taken together, the
presence of such a large number of significant characteristics in a
known pair of blue jeans precludes the possibility (or probability)
of their having occurred by mere coincidence, and they can thus
serve as a set of points of identification.
Discussion
Caution must be exercised when reaching conclusions based on
an analysis of wear patterns alone as described herein. Stray marks
on clothing such as paint or blood stains and rips or tears can be

fully expected to occur in random locations and patterns and to persist as visible features over time. FBI examiners have been making
photographic identifications based on such random features for approximately thirty years. The case described herein differs from
such identifications in that the individual wear marks used as the
basis for the identification represent features intrinsic to the denim
trousers, and do not originate from any source other than through
the manufacturing process and damage-free daily wear. Until a validation study can be conducted to establish the frequency of occurrence and the probable location of stray wear marks such as the
cross-over, “H”, or “V,” it will not be possible to quantify the significance of individual occurrences of such marks when making
comparisons. Such a study is planned, but has yet to be begun.
Furthermore, such wear marks can be expected to change over
time. As a crease or wrinkle is abraded over time the width of the
exposed white un-dyed core increases. If two ridges adjacent to one
another are subjected to enough abrasion, it is possible that these
bright features would eventually merge into a single feature. A failure to consider this possibility might lead to an erroneous conclusion. A second study is being planned in which the evolution of
denim trouser wear patterns will be documented photographically
over time.
Despite the lack of such validation studies, it should be remembered that in this and other cases the overall significance of wear
marks is not necessarily based on a quantitative assessment, but on
a qualitative assessment. A single characteristic such as a stain or
unusual tear may be sufficient if there is enough image detail pre-

Copyright by ASTM Int'l (all rights reserved); Thu Mar 8 09:01:09 EST 2018
Downloaded/printed by
Ryan Gabrielson (ProPublica) pursuant to License Agreement. No further reproductions authorized.

 VORDER BRUEGGE • IDENTIFICATION OF DENIM TROUSERS FROM FILM

621

FIG. 7—Side-by-side comparison of Spokane bank robber’s right leg (“Questioned”) and right leg of model wearing blue jeans recovered from suspect’s home (“Known”). Numbers refer to features described in text.

sent to reveal that it is unique and unlikely to be duplicated elsewhere. A determination of whether individual characteristics like
the ones discussed herein are unique or not will remain unanswered
until validation studies can be conducted. Until that time, the ability to individualize an item based on a single such characteristic
will remain a matter of opinion. However, if it should turn out that
such characteristics can be duplicated, it is much less likely that
combinations of multiple characteristics can be duplicated. The
fact that the defense in this case was unable to find any pair of
trousers that exhibited more than one or two gross characteristics
similar to the five major characteristics identified on the questioned
trousers supports (but does not prove) the hypothesis that combinations of gross characteristics may be sufficient for identification.
This leads to another caution in the conduct of such examinations. The quality of the bank surveillance images is a critical
component in this type of comparison analysis, since the wear patterns used as points of identification may be very small. If the
quality of the surveillance images is insufficient to resolve a specific characteristic, then that characteristic cannot be used as a
point of identification. Furthermore, in such a case, the absence of
a characteristic from the surveillance image due to poor resolution
cannot be used as a justification for eliminating an item. As banks
and other businesses convert their surveillance systems from film
to video systems, the likelihood of an examiner being able to reach

a meaningful conclusion in this type of case will be severely restricted. This is due to the much lower resolution provided by
video systems which are typically one-tenth that of film systems
(3). Unless the identifying characteristics under consideration are
sufficiently large, video systems cannot be relied upon to record
them with sufficient detail to permit their use in a comparison.
Hopefully, higher resolution systems such as High-Definition
Television (HDTV) and high resolution digital still cameras will
replace low-end VHS systems in surveillance systems as their
costs are reduced.
As a final caution, the importance of experience and training in
the conduct of examinations such as this cannot be underestimated.
Without a good working knowledge of the photographic process,
an inexperienced individual might misinterpret a simple difference
in perspective or a defect in the film as true differences in the shape,
size, or characteristics of a photographed object when compared
with an item of evidence. Similarly, without an understanding of
the means by which clothing is manufactured and the means by
which class and individual identifying characteristics are generated, an individual could misinterpret the significance of a given
characteristic or set of characteristics and reach an incorrect conclusion. To avoid these problems, individuals should receive extensive training prior to conducting examinations on actual casework.

Copyright by ASTM Int'l (all rights reserved); Thu Mar 8 09:01:09 EST 2018
Downloaded/printed by
Ryan Gabrielson (ProPublica) pursuant to License Agreement. No further reproductions authorized.

 622

JOURNAL OF FORENSIC SCIENCES

Summary
Wear patterns and characteristics along and adjacent to the
seams and hems of denim trousers can be used as a means of positively identifying or eliminating the trousers worn by individuals
depicted on bank surveillance film. These characteristics can have
their origin in the manufacturing process, and their visibility will be
amplified through abrasion and the resultant removal of dye-rich
outer layers. By comparing the morphology and location of such
characteristics on trousers depicted in surveillance images with the
characteristics on trousers recovered from suspects, an examiner
can conduct a point-by-point analysis as a means of attempting to
identify or eliminate the suspects’ trousers.
If the questioned image is of insufficient quality to distinguish
such characteristics, then it may not be possible to reach a conclusion. However, if the image quality is good and if a sizeable number of wear characteristics can be observed, then a unique association is possible, as demonstrated in this case. Statistical studies
remain to be conducted to assess the significance of single characteristics as a means of individualizing denim trousers.
Acknowledgments
The assistance of George Savastano in the modeling of the recovered blue jeans is greatly appreciated and herewith acknowl-

edged. The author also wishes to thank Supervisory Special Agent
Photographic Examiners Gerald B. Richards (retired), William B.
Stokes, and Thomas J. Forgas for their guidance and support in the
course of this work, as well as their thorough reviews of prior versions of this manuscript. Careful reviews of the manuscript by Photographic Examiners David A. Bonner, David B. Davies, Wm.
Grant Frensley, Christopher J. Kay, Barry R. King, Thomas
Musheno, George R. Skaluba, and Peter J. Smith are also gratefully
acknowledged. Finally, the comments of two anonymous reviewers also proved extremely helpful and are gratefully acknowledged.
References
1. Tuthill H. Individualization: Principles and procedures in criminalistics.
Salem, OR: Lightning Powder Company, Inc., 1994.
2. Avignone JM, Rielly D. Photographic analysis of bank robbery films. FBI
Law Enf Bull 1979;48(11):21–5.
3. Blitzer HL. Assessing forensic imaging options. Security J 1997;9:35–41.
Additional information and reprint requests:
Richard W. Vorder Bruegge
Examiner of Questioned Photographic Evidence
FBI Laboratory
Special Photographic Unit
Room 3449
935 Pennsylvania Avenue, N.W.
Washington, DC 20535

Copyright by ASTM Int'l (all rights reserved); Thu Mar 8 09:01:09 EST 2018
Downloaded/printed by
Ryan Gabrielson (ProPublica) pursuant to License Agreement. No further reproductions authorized.