CLPEX Chat Board

Our agency is currently preparing the infrastructure our next generation AFIS. One of the current areas of consideration is a palm AFIS. As such, I was wondering if anyone had any experience regarding palm images.

Specifically, I am curious about stitched palm images. Has anyone had any challenges to stitched images in court? Do they pose any workflow dilemmas for those in your agency responsible for capturing them? Are there common areas of distortion vs traditional methods? etc, etc.

Any and or all feedback would be greatly appreciated.

If you'd like to see a list of benefits touted by a livescan manufacturer of single vs stitched images, Clinton County PA has the White Paper for the identix 3800 here: http://www.clintoncountypa.com/Central% ... vescan.pdf

From the Identix Advertisement:

Optics(TM), cylindrical imager enables operators to accurately
capture an image from the carpal tunnel to the tips of the
fingers in one continuous scan.

From the carpal tunnel to the tips? I don't think the other vendors can make such a claim, nor would they want to

Hi Boyd,

When we were looking to buy a new LiveScan and AFIS system, I wanted one whole system from one vendor. My concern was if I bought a LiveScan from one vendor and an AFIS from a different vendor that they wouldn't work together, no matter what the vendors said. Then, each vendor would blame the other for the problems. I also wanted a LiveScan that would do palm prints that the jail folks would like and have one service contract for both pieces.

We got a Cogent LiveScan and AFIS, plus a smaller Cogent LiveScan for the Juvenile Detention Center. All the prints feed directly into my AFIS and hardcopies of the fingerprint cards print out at the State Police ID Bureau. The jail folks like the LiveScan so much better than the other machine we had (brandname withheld). Even though they have to do palm prints now, they tell me it is faster to do fingerprints, palms and writer's palms on the Cogent LiveScan than doing just the fingerprints on the other machine. Plus, the Cogent LiveScan was far less expensive than replacing the LiveScan from the other vendor with theirs that did palm prints.

On the AFIS side, I have gotten hits on latent palm prints that I may never have found by cold searching on printed cards for an hour or more. I am sold on palm print capture from the LiveScan and searching them in the AFIS. Plus, we have hooked in the local city police to share our database so they can use the AFIS. After all, their burglars are our burglars. Another city police is in the process of setting up their Cogent LiveScan for cite and release purposes and we are going to get them hooked into our database, too.

I don't know what you mean by "stitching" but it sounds like a salesman's attempt to confuse you. I would be more interested in results by other users - latent print examiners and jail folks - than what salesmen may tell you.

Kathy

We have a Spex Forensics Printquest AFIS/APIS system and I've been sold on palmprint searches for a while now. We're in the process of getting a livescan as well and if I have my way it'll be a Crossmatch ID2500 with one pass palm capture. I don't know that I would trust a computer to stitch a palm image together from multiple passes, but that's entirely personal speculation on my part. I haven't seen very many folks that scan palms yet and what their results have been.

Our agency purchased several live scan units with palm capture for our jail and other police departments in the county. We had a few vendors give us demos of their products. We found Crossmatch to be more user friendly and liked the cone shaped palm capture device better than the rollers used by Identix which can cause slippage and distortion in an image.

I have no direct experience with livescan palm capture devices, but I would caution anyone to make sure the device obtains complete friction ridge detail in the center of the palm and in the critical crease areas at the finger joints and interdigital areas. Then make sure that no one is allowed to use the capture device who has not been thoroughly trained in using it. As an examiner, I can tell you that the blank areas are always what you need. Palm scans do tend to be better than finger scans for the captured areas.

Just to bump this up a little bit, we finally got our act together and will be placing our order as soon as the PO is approved by City Hall. Weopted for a Crossmatch L-Scan 1000P on a Dataworks Plus platform. The scan resolution for palms is 1000ppi and they are awesome. It's a flat platen, but the rep brought a unit to our agency for a demo and I was able to capture my entire palm on the first try without any instruction. Their NIST Manager software is just a bonus that'll save us the time, energy and expense of filing hard copy fingerprint cards, but still retaining the ability when they are needed for court or comparisons. I can network to our AFIS and our RMS seamlessly and store mugshots, demographics and scars/marks/tattoo images along with my fingeprint records as well. The best part is we saved about $5k over the ID2500's price and got more bang for the buck. If anyone has any questions about the system feel free to shoot me a PM.

We utilize the latest generation of Motorola-BIS (Printrak). This is a rather comprehensive system with mug photo, two finger enrollment for minor charges (like DIP), or to verify that the jail is releasing the right inmate, barcode scanning of warrants, etc.

I am not sure if the palms are stitched; they probably are. The rolled impressions are not. The left & right four slaps--plain impressions are stitched--I think.

We just started building a database for palms, built mainly by livescan felony bookings. We have had some hits already--latent to palm. The system is not set up to search the livescan palms in reverse like the livescanned 10-prints.

Over the years, we have never had a livescan print challenged in court.

We nearly always have the deputy who rolled a defendant's prints testify to taking a subject's 10-P fingerprints when a latent case goes to court. About three or four years ago, during a preliminary hearing, a judge sat up and asked, "What was that term you're using?" Either the deputy or the latent examiner said, "Livescan, Your Honor." The judge said, "What's . . . livescan?" Our examiner tried to explain it to him and they had a short conversation as if they were in a restaurant or something. The judge said, "Where have I been all these years?! I never heard of such a thing!" Then he said, "Fine, let's move on!" And the case went on like nothing ever happened. The rest of us fellow examiners were there just to observe our coworker in court. It was one fo those moments when we we having trouble keeping a straight face. I was thinking, "I'm glad I'm not the expert witness today!"

The Virginia State Police database utilizes NEC. The compatability issues are minimal compared to previous years. Every state hase their own mandates and quirks on what data is to be transmitted with the prints from the local agencies to their database.

I think the one thing that needs to be addressed when exploring a purchase of this type is to sit down and map out all the requirements from all the participants. Go back to this several times to make sure you have covered all the bases. Make a large wall chart or have a large whiteboard to map it all out. Testing and testing some more is essential. We have one of the most complex workflows of any agency's AFIS/LS system, according to some Motorola engineers and project managers. Overtime hours build up quickly during tjhis process. It's not a good time to go get your blood pressure taken.

Then the next key is being able to work with the engineers from the vendors. Make sure they understand you! Make them repeat it back to you so you know for sure the understand it the way you intended. (The cycle of communications theory.)

We find that attending user's conferences is a great way to learn more about these systems. I suggest going to user's conferences of other vendors-even if you don't use their system. Both Motorola & NEC have very good user's conferences. I think Cogent has them as well.

Finally, I firmly believe one of the keys to any good livescan system is making it cop-proof and cop/deputy friendly. (I hope I did not step on too many toes with that one.) Basically, you need a structured training progam in place so that the officers are capturing the best prints possible as well as completing all the madated fields. There needs to be remedial action if their performance is substandard.

We have a systems manager that is very good explaining, in layman's terms, how this stuff works and what you need to do to make it work.

If you wish to email me, I can tell you how to get in touch with him. He may be very helpful in aswering your question.

Bill

Thanks for the input Bill. This has been a project in the making for at least two years. I'm just glad it's finally coming together. Once we got over the largest hurdle which was budget, everything else just came together nicely.
I agree that the system needs to be officer friendly. The Dataworks software has built-in quality and sequence checks and a pretty easy-to-follow graphical interface. (gotta have pictures you know! ) The AFIS side of it will be pretty painless because the livescan stores and transfers the records as a standard NIST package. When we import them into AFIS it automatically searches the ten-prints and the palms against our unsolved database. The only issue I foresee is with the agency RMS and that is because we are in the process of upgrading it as well. I have high hopes for it in any case. I feel like we did our homework.

Can someone please define "stitching"? I'm assuming it's a difference between the old and new generation Livescan images.

Old, being a scanned (stitched?) representation of the ridges and New is practically a photocopy.

I am not sure if I really understand stitching either. I hear the term kicked around and I would appreciate some expert in biometrics jumping in and explaining it. I may have to do some inquires.

But, as I understand it. stitching involves the capturing of the friction ridge images in several (perhaps dozens or hundreds) "mini-scans" while--or as--the fingers or hands are being rolled across the platen (the glass-type surface of the scanner). Then though some process, which I cannot explain since I am not a biometric engineer, this reveals the total image as one whole united image.

I am not sure what the logic was in developing this concept of capturing friction skin images, but, I think they should have thought it out further.

But, we have to keep in mind that livescans can cost upward of $65K or as little as $15K (or less, perhaps). They are not something you just pick up at Best Buy. So, in order to keep police department bean counters from balking, my guess is there is a cost factor behind the engineering of these devices.

If my conception is wrong about stitching, I hope someone jumps in and corrects me.

Wouldn't it be nice if all fingerprint examiners were dermatologists, biometric engineers and lawyers rolled into one? What a world it would be!

Bill

Stitched palm prints refer to capturing a "lower" palm print and an "upper" palm print

The system matches or "stitches" the interdigital area to confirm that upper and lower are the same hand.

When printed the lower palm and four fingers are mapped to the card with some vendors. Other vendors will print the upper and lower palm images to the card.

More importantly than the printed card is what is stored and searched in AFIS. I think we have to update our thinking from printed cards in a file to digital files in an AFIS or NIST Archive manager.

Bill,

Thanks for the clarification. I hope I didn't mislead anybody. I think I met you recently at a user's conference in California.

Perhaps you further can clarify something: In Virginia, for example, if an arrestee gives a phony name and information to the booking officer/deputy, and signs the fingerprint card, and the livescan lights-out indicates he/she has lied about their identity, they can be charged with "forging a public document." I.E., the fingerprint card is the "public document." It has been upheld in court and there is appellate case law to support these legal actions. These laws are especially useful in the area of identity fraud.

If you migrate to a paperless system in which all your fingerprint cards are archived, you essentially have no signatures, and you have no signature of the officer who captured the prints. Our deputies sign the FBI card on the back by their printed name & number. Electronic signatures like those in stores are basically worthless. There seems to be some sort of value in having someone sign the card.

I am curious how other jurisdictions handle this. We have tossed around the idea of going paperless. We are short on space. Some examiners don't like it. I am trying to maintain neutrality in the arguments. This ability to nail an identy fraud suspect is a strong argument in favor of keeping signed hard copies. However, there are warrants and other court papers that can be used instead.

I appreciate any input you or any of the other folks are willing to throw out.

Thanks,

Bill

When we went to live scan technology in 1995, one of the first problems we faced was inmates signing the wrong fingerprint cards. This was due to the lag between capturing prints and printing the cards.
When we proposed not signing the cards, I faced alot of protest from my staff. Among the complaints was that we could not prosecute people who gave a false name at booking.
Investigation revealed that due to staffing shortages etc., NO ONE was investigating, much less prosecuting inmates for lying about their name at booking. (We expect them to lie, that is why we print them)

Not only did we stop signing the cards, in 1999 we started printing the cards at the central AFIS site instead of the jail print room. This has been a much better system for us overall and was well received by the deputies at the jail. The deputy taking the prints is noted by the live scan log in. They do not sign the cards either.
This has worked for us

I work with Boyd and am on the committee planning our next gen AFIS. To clarify what he wrote in the original post, the scope of our AFIS upgrade does not include getting new Live Scan machines. We will be getting a palm matching system, but the whole question of how our Live Scan palms should be captured is moot at this point. Our current Live Scans are Identix 3800s which capture the hand from wrist to fingertip on a roller.