Fingerprint Examiners Research Articles

Face recognition accuracy of forensic examiners, superrecognizers, and face recognition algorithms

Achieving the upper limits of face identification accuracy in forensic applications can minimize errors that have profound social and personal consequences. Although forensic examiners identify faces in these applications, systematic tests of their accuracy are rare. How can we achieve the most accurate face identification: using people and/or machines working alone or in collaboration? In a comprehensive comparison of face identification by humans and computers, we found that forensic facial examiners, facial reviewers, and superrecognizers were more accurate than fingerprint examiners and students on a challenging face identification test. Individual performance on the test varied widely. On the same test, four deep convolutional neural networks (DCNNs), developed between 2015 and 2017, identified faces within the range of human accuracy. Accuracy of the algorithms increased steadily over time, with the most recent DCNN scoring above the median of the forensic facial examiners. Using crowd-sourcing methods, we fused the judgments of multiple forensic facial examiners by averaging their rating-based identity judgments. Accuracy was substantially better for fused judgments than for individuals working alone. Fusion also served to stabilize performance, boosting the scores of lower-performing individuals and decreasing variability. Single forensic facial examiners fused with the best algorithm were more accurate than the combination of two examiners. Therefore, collaboration among humans and between humans and machines offers tangible benefits to face identification accuracy in important applications. These results offer an evidence-based roadmap for achieving the most accurate face identification possible.

Proficiency testing of fingerprint examiners with Bayesian Item Response Theory1

Proficiency testing of fingerprint examiners with Bayesian Item Response Theory1

Comparing Categorical and Probabilistic Fingerprint Evidence

Fingerprint examiners traditionally express conclusions in categorical terms, opining that impressions do or do not originate from the same source. Recently, probabilistic conclusions have been proposed, with examiners estimating the probability of a match between recovered and known prints. This study presented a nationally representative sample of jury-eligible adults with a hypothetical robbery case in which an examiner opined on the likelihood that a defendant's fingerprints matched latent fingerprints in categorical or probabilistic terms. We studied model language developed by the U.S. Defense Forensic Science Center to summarize results of statistical analysis of the similarity between prints. Participant ratings of the likelihood the defendant left prints at the crime scene and committed the crime were similar when exposed to categorical and strong probabilistic match evidence. Participants reduced these likelihoods when exposed to the weaker probabilistic evidence, but did not otherwise discriminate among the prints assigned different match probabilities.

A study on the Holistic Approach for Fingerprint Examination Practice

A study on the Holistic Approach for Fingerprint Examination Practice

Certainty and Uncertainty in Reporting Fingerprint Evidence

Everyone knows that fingerprint evidence can be extremely incriminating. What is less clear is whether the way that a fingerprint examiner describes that evidence influences the weight lay jurors assign to it. This essay describes an experiment testing how lay people respond to different presentations of fingerprint evidence in a hypothetical criminal case. We find that people attach more weight to the evidence when the fingerprint examiner indicates that he believes or knows that the defendant is the source of the print. When the examiner offers a weaker, but more scientifically justifiable, conclusion, the evidence is given less weight. However, people do not value the evidence any more or less when the examiner uses very strong language to indicate that the defendant is the source of the print versus weaker source identification language. We also find that cross-examination designed to highlight weaknesses in the fingerprint evidence has no impact regardless of which type of conclusion the examiner offers. We conclude by considering implications for ongoing reform efforts.

Latent Fingerprint Value Prediction: Crowd-Based Learning

Latent fingerprints are one of the most crucial sources of evidence in forensic investigations. As such, development of automatic latent fingerprint recognition systems to quickly and accurately identify the suspects is one of the most pressing problems facing fingerprint researchers. One of the first steps in manual latent processing is for a fingerprint examiner to perform a triage by assigning one of the following three values to a query latent: Value for Individualization (VID), Value for Exclusion Only (VEO), or No Value (NV). However, latent value determination by examiners is known to be subjective, resulting in large intra-examiner and inter-examiner variations. Furthermore, in spite of the guidelines available, the underlying bases that examiners implicitly use for value determination are unknown. In this paper, we propose a crowdsourcing based framework for understanding the underlying bases of value assignment by fingerprint examiners, and use it to learn a predictor for quantitative latent value assignment. Experimental results are reported using four latent fingerprint databases, two from forensic casework (NIST SD27 and MSP) and two collected in laboratory settings (WVU and IIITD), and a state-of-the-art latent automated fingerprint identification system (AFIS). The main conclusions of this paper are as follows: 1) crowdsourced latent value is more robust than prevailing value determination (VID, VEO, and NV) and latent fingerprint image quality for predicting AFIS performance; 2) two bases can explain expert value assignments, which can be interpreted in terms of latent features; and 3) our value predictor can rank a collection of latents from most informative to least informative.

Accuracy and reliability of feature selection by Chinese fingerprint examiners

ABSTRACTThe selection of minutiae is a critical part of the analysis phase within the fingerprint identification methodology, known as analysis-comparison-evaluation-verification. This study assessed the accuracy and reliability of the minutiae selections of 92 Chinese fingerprint examiners during the analysis phase, absent an exemplar print, of the fingerprint identification process. Specifically, we measured the accuracy (trueness) of their annotation of minutiae, and we measured their reliability which is the reproducibility and repeatability in their annotations in one complex mark by using R software. We observed significant variation within inter- and intra-examiner annotations of the minutiae. We saw no statistically significant differences for the variability of minutiae annotations based on the participant's sex or years of experience.

Should Fingerprint Examiners Make More Erroneous Identifications?

This study was conducted as a response to the concerns about the consequences of latent fingerprint examinations. The goal is to determine if society’s moral values align with the current bias towards erroneous exclusion decisions over erroneous identification decisions found in latent print examinations. Subjects of this experiment were asked to manipulate a webbased visualization that reflects the tradeoffs between putting guilty people in jail and keeping innocent people out of jail. The results of the experiment were analyzed to determine the similarities and differences between the opinions of fingerprint examiners and the opinions of students and members of the general public. In practice, examiners adopt more conservative decision criteria, because they could lose their job if they put an innocent person in jail. According to the results of this study, examiners seem to have a much more liberal exclusion criterion than they actually do in casework, and the public seems willing to tolerate a higher amount of erroneous identifications in exchange for a lower erroneous exclusion rate based on their average criteria placement in the visualization. The results of this study will help examiners align their responses to those of society, and help all citizens understand the tradeoffs that can occur with shifting decision criteria. If the results of the study indicate the need to shift the decision criteria to put more criminals in jail, additional safeguards may be necessary to guard against innocent people going to jail. Thus this dataset represents a rich framework for measuring, interpreting, and responding to the values and beliefs of what constitutes a just and moral society.

Digital Fingerprint and Forensic Examination as a Technology for Cybercrime Detection and Prevention in Nigeria

Digital Fingerprint and Forensic Examination as a Technology for Cybercrime Detection and Prevention in Nigeria

Problems of use of fingerprint information in the investigation of crimes

The article examines some of the problems of using fingerprint information in the process of detection, investigation and prevention of crime. Characterized in detail the stages of fingerprint examination. The article analysed modern scientific views and ways of development and improvement of fingerprinting. In forensic science one of the earliest is traces of the doctrine of human hands, given the origin and history of use in the detection and investigation of crimes. Study of fingerprint knowledge, especially in their use of a pre-trial investigation enable to select individual events, describing the main features of the formation of this important section of forensic science and necessary activity that contributes to the detection and investigation of crimes. Obviously, fingerprint traces, as a reflection of morphological features hands and skin relief and unique features of papillary patterns, their uniqueness, individuality and immutability has long been interested in mankind. The current state of criminalistics, held in its development long way from its disparate formerly separate branches to the system of scientific knowledge before high levels of integrated implementation of the purposes of justice of scientific and technological progress, the deep development of methodological problems, tactics and methods of crime investigation. Today forensics enables high accuracy to identify the footsteps of palms or fingers. Traces of hands, which were found at the scene, is irrefutable proof that a person was at this place. One of the most critical problems is to ensure the quality of pre-trial investigation. The main purpose is scientific support for forensic practice of pre-trial investigation against crime. I must say that prompt and full disclosure of crimes is achieved during the pre-trial investigation. Using achieve natural and technical sciences, studying and summarizing good practice solving crimes, criminologists, create new, more sophisticated methods and means of fighting crime.

Organizational learning in forensic fingerprint investigation: Solving critical challenges with organizational rule construction

Abstract The present study analyses data collected from a series of developmental seminars in a fingerprint laboratory during which fingerprint examiners jointly discussed and developed their work processes, analytical methods, decision-making criteria and rules of documentation. The analysed organizational development took place in the context of moving from paper to digital documentation and from individually to collectively mastered work process. The fingerprint examiners who participated in the seminar series jointly reflected on their existing professional rules and operational practices, improvement of which was called for to facilitate organizational learning. The analysed data set consists of 10 audio-recorded developmental seminars with written documents as well as notes and decisions that were made during the seminar. The results of the study will reveal the complex ways in which the fingerprint examiners share their practical professional knowledge and collectively create decision-making criteria and rules of investigative practices so as to adapt their work practices to the changing quality requirements, evolving international standards and digitalization of research documentation.

Holistic processing of fingerprints by expert forensic examiners

Holistic processing is often characterized as a process by which objects are perceived as a whole rather than a compilation of individual features. This mechanism may play an important role in the development of perceptual expertise because it allows for rapid integration across image regions. The present work explores whether holistic processing is present in latent fingerprint examiners, who compare fingerprints collected from crime scenes against a set of standards taken from a suspect. We adapted a composite task widely used in the face recognition and perceptual expertise literatures, in which participants were asked to match only a particular half of a fingerprint with a previous image while ignoring the other half. We tested both experts and novices, using both upright and inverted fingerprints. For upright fingerprints, we found weak evidence for holistic processing, but with no differences between experts and novices with respect to holistic processing. For inverted fingerprints, we found stronger evidence of holistic processing, with weak evidence for differences between experts and novices. These relatively weak holistic processing effects contrast with robust evidence for holistic processing with faces and with objects in other domains of perceptual expertise. The data constrain models of holistic processing by demonstrating that latent fingerprint experts and novices may not substantively differ in terms of the amount of holistic processing and that inverted stimuli actually produced more evidence for holistic processing than upright stimuli. Important differences between the present fingerprint stimuli and those in the literature include the lack of verbal labels for experts and the absence of strong vertical asymmetries, both of which might contribute to stronger holistic processing signatures in other stimulus domains.

Creating the New from the Old: Combinatorial Libraries Generation with Machine-Learning-Based Compound Structure Optimization.

The growing computational abilities of various tools that are applied in the broadly understood field of computer-aided drug design have led to the extreme popularity of virtual screening in the search for new biologically active compounds. Most often, the source of such molecules consists of commercially available compound databases, but they can also be searched for within the libraries of structures generated in silico from existing ligands. Various computational combinatorial approaches are based solely on the chemical structure of compounds, using different types of substitutions for new molecules formation. In this study, the starting point for combinatorial library generation was the fingerprint referring to the optimal substructural composition in terms of the activity toward a considered target, which was obtained using a machine learning-based optimization procedure. The systematic enumeration of all possible connections between preferred substructures resulted in the formation of target-focused libraries of new potential ligands. The compounds were initially assessed by machine learning methods using a hashed fingerprint to represent molecules; the distribution of their physicochemical properties was also investigated, as well as their synthetic accessibility. The examination of various fingerprints and machine learning algorithms indicated that the Klekota-Roth fingerprint and support vector machine were an optimal combination for such experiments. This study was performed for 8 protein targets, and the obtained compound sets and their characterization are publically available at http://skandal.if-pan.krakow.pl/comb_lib/ .

Compatibility of DNA IQ™, QIAamp® DNA Investigator, and QIAsymphony® DNA Investigator® with various fingerprint treatments.

Latent fingerprint and touch DNA are the two most important contact evidence for individualization in forensic science which provide complementary information that can lead to direct and unequivocal identification of the culprit. In order to retrieve useful information from both fingerprints and DNA, which are usually mingled together, one strategy is to perform fingerprint examination prior to DNA analysis since common DNA sampling technique such as swabbing could disturb or even destroy fingerprint details. Here, we describe the compatibility of three automatic DNA extraction systems, namely, DNA IQ™, QIAamp® DNA Investigator, and QIAsymphony® DNA Investigator®, with respective to the effects of various fingerprint detection techniques. Our results demonstrate that Super Glue fingerprint treatment followed by DNA IQ™ extraction shows better effectiveness in DNA profiling. Aluminum powder dusting offers the least interference to the three DNA extraction systems above. Magnetic powder dusting, on the other hand, strongly impedes DNA recovery. Physical Developer is the most intrusive, which yields profiles with poor quality, including lower peak heights, poor peak height ratios, and poor intra-color balance. In terms of the choice of extraction method, DNA IQ™ system is recommended for sampling after fingerprint treatments, but not the two DNA Investigator systems.

Forensic Science Accuracy: Why We Know Little and How to Learn More

Forensic science – which includes such techniques as DNA analysis, fingerprint examination, and firearms comparison – plays a crucial role in our criminal justice system by helping to convict the guilty and acquit the innocent. However, our confidence in forensic science conclusions must be tempered by the odds that those conclusions are wrong. What are those odds? Nobody knows the answer because no disinterested researchers have conducted the appropriate studies in any of the forensic science disciplines. This is a serious problem because, without this information, legal decision makers cannot properly assess the validity or probative value of forensic evidence. In this paper, I examine the institutional forces and misunderstandings that are responsible for our ignorance about the accuracy of forensic science conclusions. I then recommend a new type of proficiency testing regimen (Type II proficiency testing) that is designed to measure error rates under appropriate test conditions in the various forensic subfields. Unless and until such studies are undertaken, legal decision makers will continue to fly blind when it comes to assessing the reliability of a reported forensic match.

Recent progress on fingerprint visualization and analysis by imaging ridge residue components.

Fingerprints have long been and are still considered to be the gold standard for personal identification in forensic investigations. Developing or enhancing the visualization of invisible fingerprints, so-called latent fingerprints (LFPs), remains to be the core subject in forensic science. Moreover, the past few years have witnessed a renewal of research interest in the possibility that a fingerprint can provide additional information than just identification of individuals, such as personal traits, the presence of human metabolites with diagnostic values, and the evidence of contact with explosives or illicit drugs. Fingerprint analysis has manifested itself as a research area far beyond the scope of forensics, to which not only conventional fingerprint examiners but also researchers from chemistry, biochemistry, medical science, material science, and nanotechnology fields have made significant contributions in recent years. Beginning with a brief overview of the components present in LFP residue that essentially determines which method or reagent will give the best visualization result, this paper reviews the progress since 2007 on new reagents and methods developed for LFP visualization and simultaneous detection of specific chemicals present in the LFP residue, with an emphasis on the utilization of mass spectrometry, infrared spectroscopy, nanoparticles, and immunogenic and nucleic acid reagents.

Forensic Examination of Fingerprints: Past, Present, and Future

Fingerprints have been a valuable method of personal identification for more than 100 years due to the considerable variability in fingerprint patterns among individuals. Fingerprints were initiall...

The Policy Implications of Research on Fingerprint Examination Tasks

Most fingerprint comparisons are still done by human examiners, who examine two impressions to determine the amount of perceived detail in agreement. Examiners must rely on their training and experience to determine whether the quality and quantity of detail in agreement is sufficient to warrant an identification decision, which makes their perceptual and decision-making abilities central to our understanding of the strength of fingerprint evidence. Research on latent print examiners has documented the influence of configural processing, greater working memory, and greater consistency of eye gaze among experts relative to novices. All of these lead to universally higher accuracy relative to novices. However, examiners must contend with fatigue and the problem of non-mated prints that are somewhat similar in appearance. Surprisingly, this problem only gets worse as databases increase in size. Currently, the field contends with a relatively high number of erroneous exclusions and inconclusive decisions, which may allow a potentially guilty suspect to remain free from charges. We discuss policy implications that follow directly from the research and suggest future research directions that address unresolved issues.

Discrepancies in expert decision-making in forensic fingerprint examination

The purpose of the present study was to analyse professional fingerprint examiners’ investigative practices in the context of discrepancy decisions concerning challenging latents during fingerprint analysis and identification. The participants were fingerprint experts from the Forensic Laboratory of the Finnish National Bureau of Investigation. The data were from five audio-recorded “discrepancy meetings” where two examiners were discussing the rationale and justification for their differing interpretations of challenging and distorted fingerprint evidence. The meetings were chaired by the quality manager of the fingerprint group, who also in the first author of this article. The research questions addressed were as follows: What does the examiner see in the latent fingerprints? What does the examiner actively do with the latents? How were decisions made during the investigative process? In accordance with Goodwin's professional vision framework, the results revealed how the participants used partial and limited information in making judgments about the difficult and distorted latents. The examiners appeared to be involved in active, constructive efforts, mentally, to repair poor latents by supplementing with missing information. They also highlighted various aspects of latents by colour coding as well as manipulated fingerprint images in several ways so as to make the significant patterns easier to recognize. Because the methods and practices of characterizing latents were only vaguely specified, the examiners used locally developed ad hoc practices to facilitate their investigations, ending up with different interpretations. It is concluded in the article that the fingerprint community in Finland should make strong efforts to develop the methods of fingerprint investigation and determine clearer criteria for decision making and documentation practices. Furthermore, the interpretations made by fingerprint experts should be made more transparent to the customers.

Quantifying the limits of fingerprint variability

The comparison and identification of fingerprints are made difficult by fingerprint variability arising from distortion. This study seeks to quantify both the limits of fingerprint variability when subject to heavy distortion, and the variability observed in repeated inked planar impressions. A total of 30 fingers were studied: 10 right slant loops, 10 plain whorls, and 10 plain arches. Fingers were video recorded performing several distortion movements under heavy deposition pressure: left, right, up, and down translation of the finger, clockwise and counter-clockwise torque of the finger, and planar impressions. Fingerprint templates, containing ‘true’ minutiae locations, were created for each finger using 10 repeated inked planar impressions. A minimal amount of variability, 0.18mm globally, was observed for minutiae in repeated inked planar impressions. When subject to heavy distortion minutiae can be displaced by upwards of 3mm and their orientation altered by as much as 30° in relation to their template positions. Minutiae displacements of 1mm and 10° changes in orientation are readily observed. The results of this study will allow fingerprint examiners to identify and understand the degree of variability that can be reasonably expected throughout the various regions of fingerprints.