Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror
Privacy Security Technology Your Rights Online

3D Fingerprinting — Touchless, More Accurate, and Faster 103

Posted by timothy
from the invest-in-print-eradication dept.
kkleiner writes "For all the glory it gets, the fingerprint has evolved very little in the last 60 years. They’re still two dimensional. The US Department of Homeland Security and the National Institute of Justice are hoping to change that. They've given grants to dozens of companies to perfect touchless 3D fingerprinting. Two universities (University of Kentucky and Carnegie Mellon) and their two respective start-up companies (Flashscan 3D and TBS Holdings) have succeeded. Fingerprints have reached the third dimension and they are faster, more accurate, and touchless."
This discussion has been archived. No new comments can be posted.

3D Fingerprinting — Touchless, More Accurate, and Faster

Comments Filter:
  • fingerprinting (Score:5, Informative)

    by nomadic (141991) <nomadicworld@nOSpaM.gmail.com> on Wednesday October 14, 2009 @03:05PM (#29749199) Homepage
    There is probably no scientific evidence relied upon unquestionably, that has such serious issues regarding accuracy as fingerprinting. Check this [newscientist.com] out.
    • Re:fingerprinting (Score:5, Informative)

      by Hatta (162192) * on Wednesday October 14, 2009 @03:13PM (#29749299) Journal

      Yep, here's more from Boston University [bu.edu]

      A more fundamental problem is the lack of underlying statistical evidence. The use of genetic evidence provides a good comparison. Scientists and lawyers subjected the technique, developed in 1984 and first introduced into a U.S. court in 1987, to years of scientific scrutiny and almost a decade of court challenges before it became accepted evidence. In DNA analysis, examiners identify and compare short segments of DNA--generally 13--to make a match. In addition to having established procedures for analyzing evidence, experts have calculated the odds that two people could share the same DNA in all 13 segments. These odds vary slightly based on the prevalence of certain DNA patterns among different ethnic groups but are in the tens of millions to one against two people sharing all 13 segments.

      Fingerprint examiners frequently tout the permanance and uniqueness of fingerprints, but they do not know the odds that two people could share a given number of fingerprint characteristic. With no clear rules for how much relevant weight to give to the various print characteristics, like point matches, ridge width, and the spacing of oil pores, German argues that it is impossible to attach probabilities to print identifications. Many experts believe probabilities are unnecessary since examiners would not make or confirm an identification unless they were certain of it. But when three of the most experienced FBI examiners confirm a mistake, as they did with Mayfield's prints, the argument collapses. Other print proponents argue that despite occasional human errors, the method is infallible. Critics like Simon Cole, a legal historian who has testified in many of the court challenges, rightly point out that this is a useless distinction--for whatever reason, fingerprint identifications are sometimes wrong.

      The handful of studies of fingerprints show a troubling pattern of errors. Since 1995, Collaborative Testing Services, a company that evaluates the reliability and performance of fingerprint labs, has administered an annual and voluntary test. It sends fingerprint labs a test that includes eight to twelve pairs of prints that examiners confirm or reject as matches. The pairs usually consist of complete, not partial prints, making identifications easier than the real situations examiners face. Nevertheless the error rate has varied from 3% to a dismal 20%.

    • I came here to either find or make this comment. Good job. Police and prosecutors build their careers on convictions. They have a vested interest in the public believing in the infallibility of fingerprinting. I find this paragraph from the New Scientist article to be key in understanding the controversy of fingerprinting:

      No one disputes that fingerprinting is a valuable and generally reliable police tool, but despite more than a century of use, fingerprinting has never been scientifically validated. This is significant because of the criteria governing the admission of scientific evidence in the US courts.

      • Interesting. Validating fingerprinting would be pretty trivial, given access to a large database of fingerprints.

        • by lawpoop (604919)
          Yes, but these are only of suspected criminals. What you want to know is the odds that two random people share fingerprint characteristics -- or rather, what are the chances that some innocent person gets framed because their fingerprints happen to match the actual criminal's.

          In other words, there's a selection bias in the fingerprint database.
          • by ceoyoyo (59147)

            The selection bias is only important if there's reason to believe that it has some effect on the item of interest. In this case, the dataset would be invalid if criminals have different fingerprints than non-criminals.

            Besides, there are a LOT more fingerprints in databases than those of (reasonably) suspected criminals. More than enough to show there's no criminal-related bias. As an obvious example, perhaps you've noticed that the US is fingerprinting most foreigners who cross the border?

            If all else fai

    • Re: (Score:3, Insightful)

      by icebike (68054)

      There is probably no scientific evidence relied upon unquestionably, that has such serious issues regarding accuracy as fingerprinting.

      If we were dealing with finger prints most of these issues would not arise. However, when we deal with a set of numbers in a computer file that only have to match to a certain level of precision there are way too many points for error.

      TFA lauds discusses 3D scanning and casts aspersions on pressing inked finger to card.

      I consider 3D just another source of error.

      After all, leaving a finger print involves pressure and leaves a 2D print. What would be a better comparison than another 2D print made with typic

      • by hoggoth (414195)

        Your specific points about 3D introducing possible points of failure are irrelevant. Your conjectures on the unreliability of 3D fingerprints are no better than the current blind opinion-based faith on 2D fingerprints.

        What needs to be done for 3D fingerprints it the same thing that needs to be done for 2D fingerprints, and still to this day has not been done.
        Rigorous statistical trials or studies need to be done to quantify the rate of errors in real world labs using these methods. Only then can the reliabi

        • Re: (Score:3, Insightful)

          by icebike (68054)

          No, not irrelevant.

          We all know that fingerprint identification is accurate at SOME LEVEL. The exact level has never been scientifically determined.

          Adding another process to introduce error can't be good regardless of the error rate of the underlying process.

          Fingerprints are at best exculpatory. Gross differences between two quality prints are easily detected. Even to the untrained eye.

          Close matches are difficult, and significant disagreement can ensue between professionally trained experts, even in the abs

          • by hoggoth (414195)

            Well, I still think your and my ideas about the methods are irrelevant, but consider that a direct 3D laser scan of a finger, a 3 dimensional object, may turn out to be more accurate than converting a 3 dimensional object into 2D by smudging it against a paper with ink.

    • by aj50 (789101)

      My personal anecdote about the unreliability of fingerprint scanning comes from the Science Museum in London.

      When I was there several years ago, they had an interactive creative exhibit where you could save your creations by identifying yourself using your fingerprint. I had to go through four fingers before I found one which wasn't incorrectly identified as belonging to someone else.

      While this is only an anecdote and the fingerprint scanning system was designed for a relatively unimportant scenario and pro

      • by kcdoodle (754976)
        The system you used had no human being screening the results, and what you experienced was "similar fingerprints" and probably not exact matches.

        For crimes at least, no person has ever been convicted on the testimony of an AFIS computer.

        AFIS returns "possible matches" with in a list of most similar matches to least similar. Then a latent fingerprint examiner (a real live person) determine which print (if any) is the real match.

        Also, when you watch CSI and the prints are flashing on one side of the scr
  • What comes next is the equivalent of when police drive down the street and scan license plates. You can be walking down the street and your finger gets scanned and a cop just grabs you off the street and arrests you for unpaid parking tickets.

    I'm SO glad this vital security measure will be in place.

    • by Jherico (39763)
      John Q. Public doesn't have his fingerprints on file if all he has are parking tickets. Further, the larger the total fingerprint database grows, the more computationally intensive scanning random people and trying to match them will become.
      • Re: (Score:3, Informative)

        Did you take an international flight into or out of the US lately? If so, you are in the database with all the "bad people".

        As for computational intensity, CPU cycles are cheaper than dirt, and getting even cheaper than that by the minute.

        • by Jherico (39763)
          Really? They fingerprint all international travelers now? I had no idea. Even if that were the case, international travelers is still far too small a subset to be randomly scanning fingerprints on the street.
        • Assuming that comparison of one fingerprint to another is done in constant time, then comparing a fingerprint to a database of them is done in N time, where N is the sample size. That makes comparing all the fingerprints you encounter with the database essentially an N squared operation. Faster computers can't magically handwave away algorithmic complexity.
          • Assuming that comparison of one fingerprint to another is done in constant time

            Actually, you just have to assume that comparison of one fingerprint to another is some normal distribution whose mean isn't related to the number of fingerprints in the database.

            I.e. suppose we have a database which returns fingerprint N in constant time, but the algorithm to compare fingerprints X and Y varies wildly depending on the two particular fingerprints being compared. Since this isn't related to the number of fingerprints in our database, we can still express comparing fingerprints in terms of O(

        • by N3Roaster (888781)

          Yes, two international trips earlier this year. Flying out of and then back into the US each time. Not once were my fingerprints taken as far as I know. As far as I know they only subject foreigners to that. Mod parent FUD.

        • OK, if you have EVER been printed, then you are in the database.

          But that data is subdivided into many categories. There are arsonists, murders, kidnappers, organized crime members and many other sections of the database.

          This is simply for faster searching. If you have a latent print at an arson scene, it would be faster to search the arsonists section FIRST, then if you do not get a match, only then do you bother searching the rest of the database.

          If every search searched the entire database, all seach
      • Re: (Score:3, Insightful)

        by icebike (68054)
        If John worked for the federal government or many state governments in ANY CAPACITY, they are on file. Jane Q Public has a far lower chance of having fingerprints on file simply because far fewer Janes than Johns serve in the military. As a college intern I worked for the Forest service. As soon as I had been there 90 days it was down to the cop-shop for printing. That put my life of crime on hold.
        • by Bigjeff5 (1143585)

          I worked for the DoD for about a year just a few years ago and was never fingerprinted. I was a civilian employee too, not a contractor. I did get a thorough background check though. Funny, I got one for my current private job too.

          Anyway, I think your fingerprinting was a state mandate, not a federal one.

          • by fluffy99 (870997)

            If you really had a DOD background investigation done in the last few years, then I'm pretty sure you got fingerprinted. The few agencies that do the background checks absolutely require them. What's your name, soc number and home mailing address? I'll pull up your file and see if they're present.

      • by HiThere (15173) <{ten.knilhtrae} {ta} {nsxihselrahc}> on Wednesday October 14, 2009 @04:28PM (#29750251)

        What's the rate of false positives? If you say there aren't any, I'll know you're lying.

        The correct answer is "Nobody knows, and the research to calculate it isn't allowed."

        For normal finger prints this could have been calculated decades ago, but the necessary agencies have consistently refused to permit their techniques to be evaluated. (Others have said that informal estimates show up to a 20% error rate [varies with the lab and the time period...low estimate was 3%]. I think was was being investigated was false negatives, though. I don't know the study, so I can't say for sure. This was reported to be based on voluntary cooperation of the fingerprinting labs, though, so the real numbers are probably higher.)

        (OTOH, the study reports may be someone's invention. I haven't seen it. I do know that there had been no official evaluation the last time I looked into the matter [a few years ago].)

        • by WCguru42 (1268530)

          >

          The correct answer is "Nobody knows, and the research to calculate it isn't allowed."

          I think more correctly is that it isn't profitable. I don't know all that much about finger print analysis but there probably aren't too many methods. If someone wanted to verify / invalidate the effectiveness of these methods they could but nobody will pay them to do it and so it doesn't get done. Bureaucracy at work, isn't it a beautiful thing.

          • by HiThere (15173)

            Possibly now the answer is "isn't profitable". When I looked into it in the past accurate information would have required cooperation of the FBI and various other police groups, and many were refusing to cooperate. (I'm not sure that any were so willing. The study couldn't be done, so if any were willing to cooperate, they didn't have the opportunity.)

            E.g., at that time the FBI maintained *the* database of fingerprints. But it was not available to researchers wishing to check for prints being properly r

      • Re: (Score:1, Insightful)

        by Anonymous Coward

        Unless John Q. Public has a concealed carry permit or has ever worked for a bank or other institution governed by the SEC or any other number of things. There's many non-criminal reasons for John Q. Public to end up with his fingerprints in a big government database and anyone concerned about it has a damn good reason imo.

      • Eh, we'll just build them into door handles in airports and govt buildings. Flat piece of glass on the back, slightly thicker handle to contain the camera, and you just got fingerprinted, unless you were wearing gloves.

      • by DRACO- (175113)

        Finger prints are in the cloud. Bank requires a thumb print, the Department of motor vehicles takes prints for driver's silence replacements.

  • Why are we talking about fingerprinting with these 2 words ?
    • by SEWilco (27983)
      Stop talking and come along quietly. We know you were at the crime scene because you didn't touch 3D objects.
  • by sakdoctor (1087155) on Wednesday October 14, 2009 @03:11PM (#29749269) Homepage

    Fingerprinting technology is only useful to the man, for keeping you down.
    Ever since Men in black, I have been waiting for the shiny fingerprint removing sphere.

    Where the hell is it! And where's my flying car.

    • Re: (Score:3, Interesting)

      by bertoelcon (1557907)

      Ever since Men in black, I have been waiting for the shiny fingerprint removing sphere.

      You can burn your fingerprints off on a flat heated surface, not that I tested it or anything. At least not on my hands...

      • Re: (Score:3, Interesting)

        by Thinboy00 (1190815)

        Ever since Men in black, I have been waiting for the shiny fingerprint removing sphere.

        You can burn your fingerprints off on a flat heated surface, not that I tested it or anything. At least not on my hands...

        Your original fingerprints will (eventually) grow back/be detectable.

      • by mhajicek (1582795)
        Just wear gloves...
  • that doesn't see any real improvement over the old method? Strikes me as a solution waiting for a problem. Sometimes, the good ways are the old ways. sometimes.
    • by pz (113803)

      that doesn't see any real improvement over the old method? Strikes me as a solution waiting for a problem. Sometimes, the good ways are the old ways. sometimes.

      You're missing the point by assuming that this technology will be used for forensic uses, like you see on the television. This technology is useful to identify living people (or at least reasonably intact bodies). The US Government currently collects thousands upon thousands of sets of fingerprints each day at the international borders from visiting non-citizens. My best bet is that the 3D technology is aimed at that use.

      • by Caue (909322)
        not really. I just wonder if a 3D model would be so much more precise than a 2D one; since it's only one surface. Nothing to do with forensics, just plain need. Face recognition, optical recognition, voice recognition, penis recognition, butt, leg, feet, etc. Each one sets a hole new group of options, but not any advantages over the last group.
  • The article did not say the price, unless I missed it, but I can say its going to be a hell of alot more than a bit of ink and a piece of paper. And what is the point? Fingerprints on stuff are already 2D, why do we need to check 2D against 3D?
    • by KillerBob (217953)

      False positives, and worse, False negatives.

      Because current fingerprints are a transcription of a 3D object into a 2D format, one that's often smudged, partial, or otherwise. The idea is to ultimately transfer the entire existing fingerprint database into 3D format... while you'd still be comparing 2D images from fingerprints taken from crime scenes, it would allow for easier and more accurate identification of people who are unknown, when their person (or corpse) is available.

      • Re: (Score:3, Insightful)

        by kimvette (919543)

        False positives are worse; it is better that 100 guilty men go free than one innocent man suffer.

        • Re: (Score:3, Informative)

          by KillerBob (217953)

          When talking about convictions, yes. False positives are worse. When talking about investigations, false negatives are worse. A false positive during an investigation means that you spend a little time and resources investigating and proving somebody's innocense. A false negative during an investigation means that you might let the guilty party walk free, uninvestigated, because you don't believe they're the one.

          In an ideal world, at least. :) In the real world, things are never so cut/dry as that.

          • In an ideal world, at least. :) In the real world, fingerprints are enough to get a conviction.

            There, fixed that for you.

            • by smoker2 (750216)
              No they are not. You cannot convict on fingerprints alone, there must be other corroborating evidence. The utility of fingerprinting comes in because if you can place the guilty person at the scene, it usually convinces them that the game's up and they'll either roll over or get worried and make other mistakes. But that doesn't work on an innocent person. One piece of circumstantial evidence alone cannot be used to convict.
        • by PitaBred (632671)
          Stop espousing the views of our founding fathers and toe the line of the government. "of the people, by the people, for the people" my ass.
    • Re: (Score:3, Interesting)

      by WaywardGeek (1480513)

      Every optical mouse has both a light source and a digital camera, yet they cost $20. A 3-D fingerprint scanner requires probably one extra camera. If they build 100 per year, they'll cost thousands of dollars. If they build a million, they'll be under $100.

      I started a company doing EDA and ASIC IP, but at the time, my favorite second alternative (back in 1999), was building a 3-D scanner out of 2 digital cameras and some software. I wanted to scan women so they could load a fairly accurate body shape on

    • by Ohrion (814105)
      It's useful for biometrics. Like for security, not for suspect identification...
    • Re: (Score:3, Insightful)

      by John Hasler (414242)

      > The article did not say the price, unless I missed it, but I can say its
      > going to be a hell of alot more than a bit of ink and a piece of paper. And
      > what is the point? Fingerprints on stuff are already 2D, why do we need to
      > check 2D against 3D?

      Speed and accuracy. The market is biometrics, not CSI.

    • by jhfry (829244)

      Ink and paper is going away, slowly but surely.

      I support a handful of optical fingerprint machines, that cost upwards of $5000/ scanner.

      This so called 3-D scanner is simply going to be a very precise laser range finder that scans like a barcode reader. I would imagine that it will be cheaper to manufacture (who knows what the R&D costs will be) but there are significant advantages to this method.

      They aren't really interested in the depth, only in sensing where the ridges begin and end... instead of usi

  • This might have some use in biometrics and identifying people who have already been scanned. It doesn't seem like it could be that useful forensically since prints are left on 2D surfaces.

    • This might have some use in biometrics and identifying people who have already been scanned. It doesn't seem like it could be that useful forensically since prints are left on 2D surfaces.

      FTFA: "To integrate with [the FBI's Automatic Fingerprint Identification System] database, Flashscan has special software to flatten the 3D print into 2D without cracks or stretches."

      So this system will still be useful in comparing 3D scans with 2D prints.

      • by John Hasler (414242) on Wednesday October 14, 2009 @04:05PM (#29749979) Homepage

        Right, but the software won't flatten the print quite the way pressing the finger against an object would.

        • Re: (Score:3, Insightful)

          by Joe Random (777564)

          Right, but the software won't flatten the print quite the way pressing the finger against an object would.

          The fact that they state that their flattened prints are able to integrate with the FBI database clearly means that this isn't a problem. Hell, real-life fingerprints flatten differently against different objects, so it's not like this is some new constraint, and at least the flattening process of the 3D scanner is predictable and repeatable. In short, I don't think this will be an issue.

  • UoK (Score:1, Funny)

    by Anonymous Coward

    There's a university in Kentucky? *rimshot*

  • This may be a great improvement for biomtric applications[1] but for comparing with prints lifted off objects at a crime scene you want flat prints.

    [1] Though with a touchless system it's going to be a bit harder to make sure that's a real live finger.

    • by PPH (736903)

      Though with a touchless system it's going to be a bit harder to make sure that's a real live finger.

      Real? Maybe. Live? Well there's still the bolt cutter hack.

    • [1] Though with a touchless system it's going to be a bit harder to make sure that's a real live finger.

      It's going to make it easier, not harder. You won't be able to trick it with 2D copies of the fingerprint you're trying to present, you'll have to have an actual 3D copy of the fingerprint.

      The bolt cutter hack would still work, as the other guy suggested, but it would be harder to surreptitiously copy someone's fingerprint in 3D than it is to get it in 2D (off their glass, if the movies are to be believed).

  • False Negatives (Score:2, Interesting)

    by delta419 (1227406)
    Great, so all I have to do is soak my fingers in water for awhile.
  • by John Hasler (414242) on Wednesday October 14, 2009 @03:55PM (#29749857) Homepage

    > For all the glory it gets, the fingerprint has evolved very little in the
    > last 60 years.

    Is there a type of fingerprint that has a selective advantage? I would think you'd do better with ones like everyone else's. Perhaps after 2000 generations of CSI we'll all have identical prints.

  • From TFA:

    To integrate with that database, Flashscan has special software to flatten the 3D print into 2D without cracks or stretches.

    They're able to flatten a 3D surface into 2D without stretching it? Quick, somebody notify the cartographers!

    I wonder if this could also help me peel an orange without tearing the peeling...

    • by retchdog (1319261)

      Not all "3d surfaces" are spheres... :-/

    • by Bigjeff5 (1143585)

      That's not that hard, really. It takes some care but all you really need to do is separate the peal from the orange before removing the peal from around the orange. This can be done by creating an incision down one side of the orange and working a "bubble" of peal around the orange. Once you've separated the peal, simply slide the orange through the cut you made down the side and there you go.

      Now, the hard part is flattening it out without tearing it.

      I'm sure you realize, however, when a person leaves a

  • I agree that the fingerprint has likely evolved very little in the last 60 years - perhaps hundreds of thousands of years! However, the evolution of *fingerprinting* technology is more to the scale mentioned in the story. Or, am I being too picky?
  • by Archangel Michael (180766) on Wednesday October 14, 2009 @04:19PM (#29750141) Journal

    I pick my nose before I get my finger prints done, in front of the fingerprint tech. This new development is going to cramp my style.

  • This is obviously the first killer app for the 3D laptop: http://hardware.slashdot.org/article.pl?sid=09/10/14/1739214 [slashdot.org]
  • Gummi fingers? (Score:3, Insightful)

    by MacTenchi (104785) on Wednesday October 14, 2009 @05:42PM (#29750993)

    The real question for me is, are these things less susceptible to gummi / jello fingers than 2D scanners? Seems like they would be equally susceptible, and therefore equally weak as a door lock.

  • Whose cool-aid are you drinking? Everybody who is not a total retard, knows how much of a useless security theater it is.

    For thieves there are this incredible modern device called...I think... "gloves".
    For fingerprint scanners of the current generation, you can always take the fingerprint off a mug or glass (e.g. at a coffeehouse or bar). With a simple Tesa strip. Happened exactly like that with the German interior minister.

    And no matter what, you can always just cut off his/her finger, and attach it to a s

Man will never fly. Space travel is merely a dream. All aspirin is alike.

Working...