|Year : 2016 | Volume
| Issue : 1 | Page : 60-64
Application of a Standard Procedure to Avoid Errors When Comparing Fingerprints with Their Reversals in Fake Documents
Carlos Magno Alves Girelli
Department of Federal Police, Identification Group, Espirito Santo State, Brazil
|Date of Web Publication||3-Feb-2016|
Carlos Magno Alves Girelli
Federal Police Department, Identification Group, 01 Vale do Rio Doce Street, Sao Torquato, Vila Velha - ES, 29114-105
Source of Support: None, Conflict of Interest: None
Laterally reversed (mirrored) fingerprints are difficult to detect by applying routine search procedures. One suggestion to avoid errors when dealing with probable reversals is to perform comparisons with both direct and reversed fingerprints. This simple procedure has been applied and led to the detection of two more cases of reversed fingerprint usage in fake documents. In one of the reported cases, experts found on the web the same fingerprints used by criminals in fake documents. This finding is important because it indicates that matched fingerprints do not necessarily link different criminal cases.
Keywords: Automated fingerprint identification systems, fake document, reversed fingerprint, standard procedure, web
|How to cite this article:|
Girelli CM. Application of a Standard Procedure to Avoid Errors When Comparing Fingerprints with Their Reversals in Fake Documents. J Forensic Sci Med 2016;2:60-4
|How to cite this URL:|
Girelli CM. Application of a Standard Procedure to Avoid Errors When Comparing Fingerprints with Their Reversals in Fake Documents. J Forensic Sci Med [serial online] 2016 [cited 2020 May 31];2:60-4. Available from: http://www.jfsmonline.com/text.asp?2016/2/1/60/162807
| Introduction|| |
Laterally reversed (mirrored) fingerprints are difficult to detect. Some interesting cases of reversed fingerprints developed in different substrates were reported in the literature. ,,, Girelli  presented a case in which reversed fingerprints were detected in fake documents based on the expertise of the skilled examiners. Furthermore, preponderant, in that case, were some coincidences, or luck factors, such as all documents were being analyzed by the same experts at the same time. After discussions about the difficulties associated with reversal detection and the costs and benefits involved with an accuracy of decision making, Girelli concluded that certain criminal cases deserve special attention and require specific procedures to conduct fingerprint examination. One given suggestion was to perform comparisons using both direct and reversed prints when dealing with fingerprints from probably fake documents. This simple procedure was applied by our identification group and lead to new findings of the fraudulent use of reversed fingerprints in forged documents. The aim of presenting the following case reports was to demonstrate the functionality of the procedure suggested and presented further examples in forged documents.
Fingerprints comparison process
Fingerprints have peculiarities that make possible their use as a biometric for the identification of people. Human identification through fingerprints offers a number of advantages over other techniques such as simplicity, celerity, low cost, reliability of results, and noninvasiveness collection method; and therefore has been a widely used practice worldwide for more than one century  with extensive applications for civil and criminal purposes.
In the criminal field, questioned fingerprints (unknown donor) are usually compared with one or more unquestioned prints from suspects. These are collected as standard tenprints (cards containing fingerprints from the 10 fingers) or obtained from large databases of automated fingerprint identification systems (AFIS), when available.
The AFIS is an algorithm capable of storing and comparing fingerprints (and other ridge skin impressions). In general, the existing minutiae in the fingerprints are assigned, and their positions and directions recorded in the system. Comparisons are made based on this minutiae map associated with each fingerprint. Modern processors enable fast and efficient searches of fingerprints in a database containing millions of records of tenprints and latent prints. At the end of the search, the AFIS shows a list of candidates sorted according to a score interpreted by the system as a measure of similarity with the fingerprint being examined.
According to the identification purpose and the quality of the compared fingerprints, different levels of detail may be required to reach a conclusion. A widespread examination method for latent fingerprints in the forensic science community is the analysis-comparison-evaluation-verification, which consists of analysis, comparison, and evaluation by an examiner with subsequent verification by another examiner.  This method uses a qualitative and quantitative evaluation of three levels of detail: Level 1: Refers to overall ridge flow. Level 2: Corresponds to the individual friction ridge paths, those regions containing such minutiae as end ridges, bifurcations, dots, and continuous ridges. Level 3: Details refer to ridge structures (edge shapes and pores), and their relative arrangements.  At the end of the examination, there are three possible decisions: Identification (originating from the same source), exclusion (originating from different sources), or inconclusive.
Laterally reversed fingerprint occurs when the fingerprint is rotated 180° around a vertical axis along the finger direction, from the bottom to the fingertip. The reversal can arise: (1) Unintentionally, from the contact between two surfaces, by transferring the fingermark deposits present on a touched surface to another, ,,, or (2) intentionally, by means of the use of image editing software. 
The cases of laterally reversed fingerprint are difficult to detect, because: (1) Changes in the orientation of images can cause confusion on the observer's brain, ,,,,,, and (2) standard settings of AFIS systems are not, in general, configured to process both direct and reverse fingerprints (probably due to the associated costs and benefits).
| Case Reports|| |
One person was arrested when attempting to open a bank account using fake documents. The fingerprint present in the fake identity card used by the arrestee did not match those from the person's tenprint.
The search in AFIS was performed following the suggested procedure of using both direct and reversed fingerprints. The donor was not identified, but the questioned fingerprint matched three fingerprints from two different criminal cases stored in AFIS and its reversal matched with nine fingerprints from six different criminal cases in the system. [Figure 1] shows the questioned fingerprint in the copy of the identity card. [Figure 2] is the comparison between the reversed fingerprint and another one from cases that came from the same source.
|Figure 2: Matching from comparison between the reversal of the questioned fingerprint (left) and one from a criminal case stored in automated fingerprint identification systems (right)|
Click here to view
In this case, the reversal was detected simply by means of the application of the suggested standard procedure. There was no need of exceptional personal expertise or sum of favorable coincidences.
The questioned fingerprint of this case and its reversal match those fingerprints of the case illustrated by Girelli in his work.  The current forgery was probably done by the same person arrested before in that case who admitted to having manufactured forged documents. As mentioned in that work, "the persons who presented those forgeries to the bank are gradually being identified and arrested by the law enforcement authorities." 
To obtain a passport in Brazil, people must present personal documents in a Federal Police station. Fingerprints and a photo are acquired at that time, and the passport is issued after a few days. During this time, fingerprints are processed by the identification unit. When an eventual hit with a fingerprint from the database of AFIS occurs the passports section is advised to take appropriate action.
In this case, the fingerprints of an applicant matched fingerprints stored in AFIS from a criminal record, in which the person was using a different name. When the person went to pick up his passport, he was arrested. The applicant was accompanied by another person, who admitted to also having acquired a passport using a false name, and thus was arrested.
The arrestees held checks, credit cards, and various documents with false names. Even the car they were using was acquired on behalf of a different person, by means of a vehicle leasing from a bank using false documents. The passports would have been used to travel to another country to continue with criminal activities.
The fingerprints present in the fake documents did not match the fingerprints of the applicants. Following the suggested standard procedure, direct and reverse fingerprints were processed. The donors were not identified. One of the questioned fingerprints matched the fingerprints from different criminal cases in three states of Brazil: Espirito Santo (ES), Rio de Janeiro (RJ), and the Rio Grande do Sul (RS). Those cases were already linked in the system. The reversal of that fingerprint matched fingerprints from the cases of the state of Acre (AC) as shown in [Figure 3].  Until that moment, the cases from Acre were not linked to the others.
|Figure 3: Reversal of a questioned fingerprint (left) and its correspondent from a criminal case of the state of Acre (right)|
Click here to view
The first assumption of the fingerprint experts was that forgeries could have been produced, most likely, by the same person since both direct and reverse fingerprints came from the same source as deduced in the previous case. However, the state of Acre is very far from the others, as can be seen in [Figure 4]. This long distance does not rule out that the crimes were committed by the same criminals but makes this hypothesis unlikely.
|Figure 4: Brazilian map showing distances between states where direct and reversed fingerprints have matched|
Click here to view
Instead of believing that the forgeries were produced by the same person experts supposed that the fake documents were manufactured by independent fraudsters. To verify this hypothesis, they carried out a search for those fingerprint's images on the web considered to be the most likely place of the source. Surprisingly, in a few minutes they found a lot of web pages exhibiting those same fingerprints. [Figure 5] [Figure 6] [Figure 7] show some examples. ,,
|Figure 5: Webpage  exhibiting the same fingerprint used by criminals in fake documents|
Click here to view
|Figure 6: Webpage  exhibiting the same fingerprint used by criminals in fake documents|
Click here to view
|Figure 7: Webpage  exhibiting the same fingerprint used by criminals in fake documents|
Click here to view
There is no more reason to believe in the correlation between those cases. Their matched fingerprints are widely used and are accessible for everyone on the web. The hypothesis of independent fraudsters seems to be more robust than that of a countrywide criminal organization.
| Conclusion|| |
It is difficult to detect laterally reversed fingerprints by applying routine search procedures. The default configurations of the AFIS used by the Federal Police Department of Brazil do not automatically consider the possibility of reversals. The AFIS has a tool that allows for the reversed latent prints to be used, but it is an optional configuration and is rarely utilized. This is probably because the history of reversals has not been justified, so far, the inclusion of this routine in standard processing, which would mean an increase in workload and in response times.
The search for both direct and reversed fingerprints when dealing with likely fake documents is a possible solution that has already presented results. The simple application of this procedure led the examiners to detect the use of reversals without the need for any additional ability or luck. For this reason, we recommend the application of this standard procedure by other police agencies.
Moreover, the search on the web can be useful to find fingerprints used in fake documents when the questioned fingerprints were not found by all the other conventional searches.
The author would like to thank the fingerprint expert Paul Allan Alvarenga for the work performed with reversed fingerprints and discussion about the results, and Paulo Alves for help with the language.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Lane PA, Hilborn M, Guidry S, Richard CE. Serendipity and super glue: Development of laterally reversed, transferred latent prints. J Forensic Identification 1988;38:292-4.
Saviano J. The Significance of using level 1 detail in latent print examinations. J Forensic Identification 2003;53:209-18.
Kershaw MH. Laterally reversed. J Forensic Identification 2000;50:138-40.
Czarnecki ER. Laterally inverted fingerprints. J Forensic Identification 2005;55:702-6.
Girelli CMA. Laterally reversed fingerprints detected in fake documents. J Forensic Identification 2015;65:1-17.
Champod C, Margot P, Lennard C, Stoilovic M. Fingerprints and Other Ridge Skin Impressions. Boca Raton: CRC Press; 2004.
Fehn U, Streicher M, Ettlinger G, Brown JV. Confusion of laterally inverted mirror-images: A relationship to brain anatomy? Cortex 1985;21:91-110.
Jordan K, Heinze HJ, Lutz K, Kanowski M, Jäncke L. Cortical activations during the mental rotation of different visual objects. Neuroimage 2001;13:143-52.
Rossion B. Picture-plane inversion leads to qualitative changes of face perception. Acta Psychol (Amst) 2008;128:274-89.
Milivojevic B, Hamm JP, Corballis MC. About turn: How object orientation affects categorisation and mental rotation. Neuropsychologia 2011;49:3758-67.
Lawson R, Jolicoeur P. Recognition thresholds for plane-rotated pictures of familiar objects. Acta Psychol (Amst) 2003;112:17-41.
Thompson P. Margaret Thatcher: A new illusion. Perception 1980;9:483-4.
Busey TA, Vanderkolk JR. Behavioral and electrophysiological evidence for configural processing in fingerprint experts. Vision Res 2005;45:431-48.
Girelli CMA. Detecção de impressões digitais revertidas em documentos falsos. (Detection of laterally reversed fingerprints in fake documents). Revista Brasileira de Ciências Policiais 2014;5:11-29.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7]