Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Ideas for making an image sensor

  1. Apr 18, 2017 #1
    I know it may sound ridiculous, but I've always thought of making a simple homemade image sensor. Is it possible to make a primitive image sensor using many photocells or other electronics parts? or it's IMPOSSIBLE AT ALL?

    I'd like to know if there are DIY methods for making an image sensor.
  2. jcsd
  3. Apr 18, 2017 #2


    User Avatar
    Staff Emeritus
    Science Advisor

    Sounds feasible. You could place an array of photocells near the focal point of a lens system and read out the voltage on each cell. It just won't give you a very good image.
  4. Apr 18, 2017 #3


    User Avatar
    Gold Member

    I have done it using a mechanical scanning device and using a photo transistor or a photo multiplier as the pick up device. This produced a 32 line picture at 12.5 fps which I could view mechanically or by digital conversion to 625 line analogue system.
    If you use individual photo cells in a grid, there is a huge problem with the quantity of devices required and the method of switching.
    These issues were being looked at in the 19th Century when the challenge of devising a practical TV system arose, and Paul Nipkov suggested scanning as a means of avoiding thousands of photo cells.
  5. Apr 19, 2017 #4
    Thanks. How about the optical part? Is there any projects or videos on this?
  6. Apr 19, 2017 #5
  7. Apr 19, 2017 #6

    Attached Files:

  8. Apr 19, 2017 #7
    Good question/ If one goes back in history they also had that same problem, hence the bright lights, and sweaty look of actors with makeup.
    Makes me wonder now if the makeup was not there to cover up "blemishes" ( High Definition ) , but to give a more reflective surface to the skin of a person.
  9. Apr 19, 2017 #8


    User Avatar
    Gold Member

    In practice, a big lens is required, and the photo detector must be in a dark box and may require a condenser lens in front of the detector. If the disc is say, 50cm diameter, with 32 holes, then sufficient pick up for ordinary lights can be obtained by using a photo multiplier tube type 931. Results have also been obtained with photo transistors and photo voltaic solar cells, sometimes using discs of about 25cm. Another web site of interest is nbtv.org.
    The disc can also be used for simultaneous display, using some white LEDs with a diffuser panel.
  10. Apr 20, 2017 #9


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    A bit off topic but my father was a radio engineer involved in the early days of Television. The early black and white cameras had a strange response to colour so the actors had to wear strange coloured makeup including blues and blacks to "look right" on a B&W TV...

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted