Thursday, September 27, 2012

Apple Acquires TFT Image Sensor Patent

PatentlyApple reports that Apple has purchased TFT image sensor patent US7,773,139 titled "Image sensor with photosensitive thin film transistors". The three Oregon-based engineers that invented this process are Willem den Boer, Tin Nguyen and Patrick Green.


  1. I've seen a presentation at ESSCIRC by Philips so many years ago. Is it the same?

  2. Adding an image sensor to a flat panel display (i,e, the same size as the display) so it would need contact or projection to work. A figerprint sensor? A written page to scan sensor? A touch screen fingertip detector (that's optical not capacitive)?

    Anstract: An image sensor array includes image sensors having photo TFTs to generate photocurrent in response to received images. The photo TFTs each have their respective gate electrodes and source electrodes independently biased to reduce the effects of dark current. Storage capacitors are coupled to each photo TFT and discharged upon generation of a photocurrent. Each storage capacitor is coupled to a readout TFT that passes a current from the storage capacitor to a data line. The photo TFT may be disposed above the storage capacitor to increase the exposed surface area of the photo TFT.

    The three guys were (are) originally working for Planar Systems, Beaverton, OR (a diplay manufacturer) the original "Original Assignee".

    Application number: 10/825,92:

  3. The impression I get from this patent is pretty similar to the one I got from the Nokia graphene-based image sensor patent material noted on the blog a few weeks back.

    In the latter case, the claims basically described a graphene active pixel sensor. It didn't seem non-obvious to me that if you had a graphene photodetector and graphene field-effect transistor, you might combine them into a pixel with pre-amplification. It's not that if you made a working device it shouldn't deserve patent consideration. It's just that the language made me feel like the applicants were rushing to grab broad ownership rights based mainly on shoving something through the PTO that would be unassailable due to the expense of contesting it as an issued patent. I didn't see what was new, useful, and non-obvious in the claims.

    The independent claims in the acquired patent basically describe a pixel with a "substrate" on which are a "photo TFT" and a "storage capacitor", this pixel in an array, and the pixel with a "readout TFT". As above, I see it as obvious that if you have a photo TFT that you can integrate with capacitors and switching and amplifying TFTs, you could make an image sensor array essentially identical to a CIS array. Also as above, I failed to see what was new, useful, and non-obvious about the claims.

    1. Its obvious that if you have a CCD imager, that you can also have a CMOS image sensor (i.e., having a transistor in each pixel). So why are there so many patents on CMOS imagers I wonder?
      Extrapolating, its obvious that since we have fundamental laws of nature, everything that we percieve today is possible. What is so non-obvious about that? BTW; please tell me one invention that is "non-obvious"?

    2. RPK-

      I'm not sure whether your position is that nothing is obvious or that everything is.

      Coming up with an idea that is new, useful, and non-obvious is hard. That something is useful is usually pretty easy to demonstrate (but not always - see below). Novelty is more challenging to show, especially as so many people are doing so many things in so many areas. There is a vast amount of material to consider, but there are also efficient digital tools for sorting out the wheat from the chaff. That something is non-obvious can be very difficult to demonstrate, because it requires putting one's self in the perspective of "one skilled in the art" and justifying in context that the idea would be overlooked.

      I will offer for your "please tell me one" consideration the classic and classical example of the negative feedback amplifier. The operation of such devices is obvious now, but at the time of invention those "skilled in the art" were by and large interested in maximizing amplifier gain, so doing something that intentionally reduced the gain was not something they would do. The negative feedback wasn't prima facie useful to the experts, so it took the inventor a really long time to get a patent.

      Having that "Eureka!" moment with a new, useful, and non-obvious idea is kind of like touching grace. I've had it a few times with patents, and also in other areas, notably in a sport I play (where I'm hoping to parlay my innovation into a try-out for my country's Olympic team in a few years) and even the other day jury-rigging a steering-capable drag-line for my son's go-kart when the batteries ran out in the field.

      Getting back to the patents here, I would say both of them are vulnerable to attack. The conventional wisdom in getting a patent is to have very broad independent claims, with narrower dependent claims so that some claims remain should the independent claims be knocked down. I see that strategy in these patents. Ultimately, though, the actual invention should be clearly described in at least one claim. I don't see that here. What I see are broad (and to me not non-obvious) independent claims with obvious narrowing limitations that leave the dependent claims still quite broad. The resulting weakness, in my opinion, is that one need only focus on undermining the independent claims in order to eliminate all of the claims. And there, I'd say the prospecting is promising.

  4. Samsung and Microsoft alreay have such a product called surface.

  5. Any idea as to how much apple may have paid for this patent?


All comments are moderated to avoid spam and personal attacks.