Tuesday, March 24, 2009

Hitachi Invented Image Sensors: History Lesson

Tech-On has an article on origins of image sensors inspired by Caltech's lawsuit against Sony, Panasonic, Canon, Nikon, Olympus and Samsung. It sounds like the defenders claim that APS were actually invented by Hitachi, rather than JPL. It's interesting to see what would be the court verdict on this.

Update: Electrons and Holes blog adds few more history bits on image sensor's early years.

10 comments:

  1. Pity that the reporter does not give more details on Hitachi's claims. My info goes back to NASA in the 60's:
    http://electronsandholes.blogspot.com/2009/03/patents-history-and-origins-of-cmos.html

    ReplyDelete
  2. Thanks for the interesting link. I'll add it to the frontpage.

    ReplyDelete
  3. Just a quick scan and I come up with pat #4,473,836 from September 25, 1984 by none other than Savvas Chamberlain. Look at claim #6.

    MR

    ReplyDelete
  4. Yeah, nice non-linear CCD patent. Nice reset FET in the pixel. I heard the DALSA guys used to call this the "Dinosensor" in secret.

    The first active pixel sensor I know of was Peter Noble's 1968 proposal, referenced in Fossum's 1997 Trans ED review paper, along with alot of other history incl. Hitachi, Olympus and others. Good reading if you care about history of image sensors. Or you can learn about the Japanese version of history in the Tech-On rag if that is where you want to get your history lessons.

    ReplyDelete
  5. Interesting discussion, and I would definately also add Imec to the list of originators. Two examples:
    1. The good old FUGA, still being marketed by Cypress. The official description:
    "Name of FillFactory's family of random addressable, logarithmic response, image sensors. Mainly intended for industrial and automotive vision. The named was given by Bach fan Bart Dierickx when IMEC management demanded something less arcane than "XYW", the latter being the first FUGA-like CMOS APS IMEC made back in 1986. Incidentally, FUGA means 'flight'. A real premonition if you take into account that the FUGA15 was to become the first CMOS APS in space."
    2. The CMOS image sensor developed around the same time, but this time on high resistivity silicon, with a different architecture, and for high energy physics experiments. The sensor returns the pixel address for a hit, and a measure for the corresponding particle energy.

    ReplyDelete
  6. CCDs, MOS Imagers, AMI, CMD, BCMD, XYWZ F.u. image sensors etc were all around before or at the same time as JPL working on their CMOS APS. JPL was laughed at first by most people in CCD industry like CHamberlin and Thuewisen but JPL's idea works better than anyone elses and now everyone uses CMOS baseline with small CCD in each pixel with direct pixel readout and on chip circuits and CDS and DDS and all that stuff. No one did that before and now it is millions of sensors per month. I think JPL has a point. Now everyone tries take credit for being originator.

    ReplyDelete
  7. What does it matter who was first. We all contributed and we all make our living based on what Theuwissen, Dierickx, Vanstraelen, Fossum, Chamberlain, and many others, gave us. Should we not be working on improving the future, instead of wasting time in court, trying to spice our finances in a period of crisis.
    There are patents, and patents should be respected. The existance of a patent is however no proof for being first.

    ReplyDelete
  8. The true father of CMOS is Peter Noble from the UK. He and his team developed the first MOS-based sensors with passive 3T pixels, active 4T pixels, Capacitive transimpedance amplifier buffers, shared reset line for adjacent rows, etc. See the IEEE literature in late 1960's up to 1972.
    We should all give credit where credit is due, especially since JPL has lost its common sense.

    ReplyDelete
  9. As a historical footnote, JPL/Caltech was turned on to the emerging CMOS technology as a result of its work with Rockwell Science Center. JPL required RSC to use Reticon CCDs for some hybrid focal plane arrays as recommended by Janesick. The Reticon CCDs were very hard to drive, needed finicky controllers, and used direct injection input. Alternative hybrid FPAs using a CMOS readout with active amplifier per pixel sold JPL that CMOS was the way to go and the rest is history.

    ReplyDelete
  10. Wow, there are so many truths twisted up with irrelevant facts and outright fallacies that even Goebbels would be proud of some of these anonymous posters. For example, the RSC comment is almost totally backwards. You might as well name RCA as the father of all CMOS image sensors since they invented CMOS circuits.
    Well, I should not comment in detail nor will I comment again, but readers should know that most of thse anonymous comments are just silly and generally inaccurate. (Except for much of the Peter Noble comment)
    -EF

    ReplyDelete

All comments are moderated to avoid spam.