UV Ink Detection?

Discussion in 'The Projects Forum' started by sexy-trousers, Mar 13, 2008.

  1. sexy-trousers

    Thread Starter New Member

    Mar 13, 2008
    2
    0
    Hi, I know very little about electronics and circuit building, so please bear with me here... I'm curious to garner your opinions on whether you think its possible to create a simple, low-cost circuit for detecting ultraviolet ink - either detecting patterns or even simply detecting how much UV ink is on a document.

    Digging around on the net I am guessing the simplest method would be to use some sort of photo detector in a photovoltaic mode and determine the amount of UV radiation based on the voltage generated. This won't detect shapes etc. but it may be simple enough for what I need - does anyone know of any existing sample circuit diagrams that would accomplish this, or does anyone have any suggestions on how to do this? Detecting patterns would be best, but any suggestions are more than appreciate.

    Thanks, TJ
     
  2. thingmaker3

    Retired Moderator

    May 16, 2005
    5,072
    6
  3. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,677
    900
    One problem with an all-or-nothing approach to detect fluorescent ink is background fluorescence. There are a variety of agents, search on calcoflor white and whiteners, that are added as brighteners. Thus, some paper will have a strong background fluorescence while other paper will not.

    I think you will need to scan, maybe at low resolution, to detect signatures and messages on that background. John
     
  4. Søren

    Senior Member

    Sep 2, 2006
    472
    28
    Hi,

    What exactly ae you trying to accomplish? Might be easier to provide a hint or a solution if you tell.

    UV-inks have very different fluorescense, some give off much light, others very little and depending on the wave length of your UV light source and the peak wave length of your detector etc., you will probably get false readings most of the time.
     
  5. sexy-trousers

    Thread Starter New Member

    Mar 13, 2008
    2
    0
    Basically, I want to determine if its feasible to uniquely identify a page based on a marking, symbol or any other means without visibly altering or tagging the page. I was thinking of using UV inks as the marking agent, I'm thinking using some sort of scanning technique would be inaccurate as it could be affected by the angle of the scan, lighting conditions etc. I don't know if there is some sort of magnetic ink or other technique that I could use to accomplish the same goal, but thats general concept at least. Cost is important, so scanning the page as an image is unlikely and so is rfid - any ideas?

    Thanks for your help... TJ
     
  6. Søren

    Senior Member

    Sep 2, 2006
    472
    28
    Hi,

    Magnetic ink, as I know it, is dark brown, grey or black(ish).

    A cheap web cam might be useable for scanning and if a number of dot's are used, they could be counted by software after some image manipulation.

    A bar code could be printed in UV-ink, and read with a CCD-scanner if the red LED's are changed to UV-LED's - might take some adjustment of detection thresholds and perhaps invertion of the signal, since bar codes are read as dark on light, while you would be trying the exact opposite (the amount of UV-conversion of the paper will be defining the attainable maximum contrast ratio).

    If you don't need to make it machine readable, a "spy kit" from your local toy store (or science toy store) might be just the ticket, as I recall some pens with ink that were only visible with polarising sunglasses (also used by some poker players to mark the playing cards) - just write the page number in large digits (I assume just reading the page number with your own eyes at a normal reading distance is impossible?).

    Which method you need depends a lot on the stuff like:
    Reading distance.
    Ambient light when reading.
    Whether it has to be done secretly.
    How fast it should be done.
    Whether it can be a permanent setup or not.
    How you are going to get, or use, the results.
     
Loading...