AFIS Accuracy and the Unidentified Latent Image

In this Blog I wanted to talk about something that would be just as useful to a prosecutor as to a defense attorney. I decided to talk about the automated fingerprint identification system and peoples assumptions of AFIS systems accuracy and potential case impact If you have latent image(s) in a case that are not identified and have been searched in AFIS, can you be sure the individual is not in the system? The answer unfortunately is no. In Omaha a couple years back a latent examiner made a latent identification on a homicide from an AFIS search, I think it may have been the 4 th. search of the system before it produced the correct suspect. The examiner understood that in some cases the latent needs to be entered and searched more than once. So if there are unidentified latents in a case, a single search of the system may not produce the identification. It would seem that in cases going to trial, the attorney would want to insure the unidentified latents be entered more than one(1) time to insure that the latent is one that does not have a match in the database What are some of the causes of the system not producing the individual?




The common assumption of the court, attorneys, jurors and many latent examiners is; AFIS systems are close to 100% accurate, so one search should suffice. Most systems today will have an accuracy less than 75 t0 80 % for latent searches. Typically the vendor will produce an accuracy 75 to 80 % if the latent image has 15 points and the target has the same 15 points. The other thing to remember about these tests are the vendor is using a small database that may not accurately reflect the a larger database of the live system. What can add to the confusion is the tenprint to tenprint accuracy which for the FBI IAFIS is 98% and most state systems probably 95%.



The other factor to remember is the system will produce a number of very good suspects which can move the suspect down the list or when dealing with poor quality images an examiner may miss the identification.

Another factor to consider, minutiae placement which is critical to any AFIS search. Minutiae can be selected by the system and manually edited by the examiner or manually selected by the examiner. Anecdotal evidence would suggest manually placing the minutiae will produce a more accurate result. Studies have shown that an examiner may vary slightly on minutiae selection the 2nd. time they see a latent print (table 3).

The results of any of the above could lead to a wrongful conviction and individual not being interviewed who may help in the investigation or an element of doubt in a jurors mind. To insure a proper AFIS search the prosecution should insure that latents that may impact a case if unidentified are searched more than one (1) time when they have their meeting prior to the court date. The defense attorney on the other hand would be the fail safe, insuring that the latent was searched multiple times.

Bob McAuley Dir. Operations/Training Forensic Biometric Identification Solutions LLC.

Cognitive issues in fingerprint analysis:Inter-and intra-expert consistency and the effect of a ‘target’ comparison

This study and paper published by Dr. Dror; documents minutiae selection during the analysis stage of a latent print. The analysis of the latent is critical because it sets the stage and the parameters for comparisons and decision making. The report is a good read for anyone dealing with latent fingerprint identification. The paper can be accessed at:


https://docs.google.com/viewer?a=v&pid=explorer&chrome=true&srcid=0B3bls8DXy0ybYTk1MTFjZTctZDEwNC00YzMzLTg5NmMtNWNiZjk5OTNlMTVk&hl=en&authkey=CMrUwuwI

Dr. Dror's summary and conclusions present logical suggestions on how to help alleviate the issue's, however with today's budgetary restraints, agencies will be hard pressed to implement all the suggestions. If the information contained in the paper on minutiae selection is incorporated into the fingerprint training programs for new examiners as well as continuing education for experienced examiners it can go a long way toward improving minutae selection on the initial analysis.

Bob McAuley Dir. Operations/Training

Forensic Biometric Identification Solutions LLC.

Fingerprint: The Foundation for quality latent comparison work ...

Fingerprint: The Foundation for quality latent comparison work ...: "Not all latent examiners are trained using standard training objectives that insure the latent examiners competence. The competence of the l..."

Bob McAuleyDir. Operations/TrainingForensic Biometric Identification Solutions LLC.

The Foundation for quality latent comparison work is laid in the initial training.

Not all latent examiners are trained using standard training objectives that insure the latent examiners competence. The competence of the latent examiner in many states is left to the local agency and courts may or may not fully understand the training received by the latent examiner. With the weight given fingerprint evidence in court cases, it is critical the examiner be competently trained. As the title indicates the foundation for comparison work is laid in the individuals’ initial training. As indicated earlier there is no standardized training program, license or training accreditation. The courts and attorneys often assume that if an agency meets accreditation standards that the latent examiner has had a better training program and will produce consistently more accurate work, which is not necessarily correct. The American Society of Crime Laboratory Directors (ASCLAD) Laboratory Accreditation Board only requires an agency to have a documented training program and follow it. So it is up to the agency to determine what the training curriculum will contain, there are no standards or guidelines. Many agencies rely on the Scientific Working Group on Friction Ridge Analysis, Study and Technology (SWFAST) suggested guidelines for training to competency which come the closest to a national standard. The guidelines are good but lack the specifics that would help an agency develop a standardized training program. The other problem today is that often the unit directors are proficient in other forensic disciplines and may not have the understanding of latent fingerprints that would be required to oversee a latent training program. In a recent study 72% of agencies surveyed indicated they adhered to training guidelines published by SWFAST another 21% did not use the guidelines and 7% did not know if they used the guidelines. In the survey the respondents were asked questions which would indicate if SWFAST guidelines were actually being followed, the answers indicated that only about 25% of the agencies actually followed SWFAST guidelines in their training program. The Federal agencies did best followed by the state with local agencies at 8% following SWFAST training guidelines. The other point I want to make is that vast majority of latent fingerprint cases taken to court are done at the local agencies, in the study 62% of local agencies had no written training program in place.


The other point brought out in the study was the discrepancy in length of the training program, the training ranged from 6 mo’s to 48 mo’s. SWFAST guidelines suggest a “Minimum: One year of full-time latent print work with the majority of the time spent on the analysis, comparison, and evaluation of impressions. Recommended: Two or more years of full-time latent print work with the majority of the time spent on the analysis, comparison, and evaluation of impressions.”

The County Attorney/District Attorney office should insure the local agency have a training program that will insure competent and accurate latent comparison work.

Bob McAuley Dir. Operations/Training Forensic Biometric Identification Solutions LLC.

School buses test fingerprint scan

School districts are turning to high-tech solutions such as fingerprint scans to track kids on school buses and keep them from getting off at the wrong stops. Unlike the automated fingerprint identification systems used in law enforcement the fingerprint images are not stored. The fingerprints are converted into a series of numbers that cannot be used to re-create a fingerprint. These systems insure privacy and since you are dealing with such a small database, accuracy is not compromised. The system will insure that an accurate picture of when and where the children boarded and left the bus. The system also insures the children get off at their designated stop with an alarm that will sound alerting the bus driver if the child tries to depart at the wrong stop. The equipment today is small and can easily be adapted to a bus and not add any additional time requirements boarding or departing the bus. The drivers are not required to do anything with the equipment everything is automated.




Some additional thoughts on what could and should be incorporated in a school bus biometric system:



Consideration should be given to use biometrics to make the school buses more secure, requiring a fingerprint in order to enter and start the school bus. Most parking bus parking lots have little or no security, requiring drivers and staff to utilize a fingerprint to allow access to the area would certainly be beneficial. Most modes of transportation have seen an increase in security but the school bus industry has not and actually is quite vulnerable. Utilizing biometrics would be a good 1st. step in securing everyone’s safety.



A biometric system could be adapted to monitor how well the driver meets schedule requirements or a way to document maintenance on the vehicle.



Bob McAuley Dir. Operations/Training Forensic Biometric Identification Solutions LLC.

Utilizing Lesser Quality Latent Images to Exclude an Individual.

The court and attorneys are made aware when a latent image in a case is identified to the defendant or victim. The latent is often the critical piece of physical evidence in a case. What the court and attorneys may not be aware of are the latent images that are not verifiable or categorized as no value. I want to discuss the unverifiable or no value images, images that lack the quantity and clarity of detail for identification. What you may not know is the unverifiable or no value images can also play a significant a role in the identification process. These lesser quality images can sometimes be used to exclude an individual based on the one discrepancy rule. Simply a single difference in appearance between a latent print and a known fingerprint must preclude identification unless the examiner has a valid explanation for the difference. Distortions are not considered a discrepancy and not a basis for exclusion. The “Exclusion” is supported by the theories of biological uniqueness and permanence, probability modeling, and empirical data gained through more than one hundred years of operational experience.


Currently there is no uniform policy/procedure on how CSI units handle the collection and evaluation of latent fingerprint evidence. In some units the evaluation of the latent may be done at the crime scene other units may do it back at the Unit Lab. The evaluation of the latent in the controlled environment of the Lab. should produce a more accurate result. What is critical to an accurate analysis are the agency policy/procedures and qualifications of the full time latent examiner making the analysis. The designation of unverifiable or no value images may result in the images being discarded/destroyed that could be used to exclude an individual as the donor of the latent image. As is obvious, the more experienced latent examiner should produce a more accurate and uniform result. In the 80’s it was possible for the crime scene tech. to determine a latent prints usability in the field, but with today’s image enhancement tools, latent images require a a more experienced latent examiner make the initial evaluation on value.

In July of this year the International Association of Identification dropped its ban on qualified conclusions, and opened the door for future testimony based on probability models involving finger/palm print comparisons. In the not too distant future you can expect to see the unverifiable images coming into court as probable identifications. This new policy just amplifies the need to insure the initial latent image analysis is uniform and accurate insuring critical evidence is available to the court. This is situation is one where if the agency policies/procedures are not updated/maintained resulting loss of evidence could impact a prosecutor as badly as a defense attorney.



Bob McAuley Dir. Operations/Training Forensic Biometric Identification Solutions LLC.

Latent Finger and Palm Print Verification Techniques

Our training course is open to everyone, you do not need to work in law enforcement to take our training. I see many advantages to a Law Office that would like to be able to determine when an expert fingeprint examiner is required since our course at is so reasonably priced. The training course adhere's to the Scientific Working Group Fingerprint Ridge Analysis, Study and Technology training guidelines for latent examiners. Goals expected from this training include:

Ability to properly conduct a comparison and render a proper and accurate conclusion.

Understanding the use of fingerprints, palmprints in criminal and civil applications.

Understanding Friction ridge examination (analysis, comparison, evaluation, and verification).

Ability to analyze friction ridge details to determine the value for comparison.

Recognize/utilize ridge flow, scars, creases, and other detail supporting examination.Recognize and properly determine,the area from which the friction ridges originated.

Properly analyze latent impressions,understand pressure distortion,slippage, and overlays.

Documentation of a latent examination.Ability to produce a court exhibit.
Understand Daubert.

The Benefits to an Office are many:
Course is reasonably priced at $378 total
The student is able to progress at their own rate
No per diem or travel costs.
Comparisons in question can be explained using electronic exhibits, similar to court exhibits
Sincerely

Robert McAuley

Fingerprint Biometric Identification Solutions LLC.
Omaha, NE
ph: 402-321-3933
alt: 402-707-9013
FBIS @cox .net