The new sensor is said to have <1.2e readout noise, >55% QE, dark current of <10pA/sq.cm at 20C, provides greater than 88dB of intra-scenic DR and produces images at rates up to 100fps in rolling shutter mode or 50fps with global shutter, booth at full resolution. PRNU is specified at <3% RMS, full well is 30,000e-, image lag is 0.1% of maximal output, MTF at Nyquist is 0.4.
Incidentally, Laser Focus World published a comparison article on low-light imaging with ICCD, EMCCD and sCMOS sensors, written by Michael Buchin, president of Stanford Photonics. The author seems to favor sCMOS approach.
Hello all, i apologize for posting a comment not related directly to this particular topic. I am trying to learn about CMOS image sensors and the design. I was wondering if i am correct in classifying CMOS image systems as array based sub-systems similar to memory design (SRAM etc). Since in SRAM for example, we have array of cells, while in Image sensors, we have active pixels.
ReplyDeleteTrue, both RAM and image sensors have array structure. However, the whole underlying technology is different.
ReplyDeletehaving read the article, I would say that the author (Michael Buchin) seems to favor the ICCD approach over all others.
ReplyDeleteOnly at the end does he say that sCMOS may eventually be able to compete with the EMCCD once additional performance ehancements become reality for the part.
Indeed, the article goes into lengthy promotion of ICCDs made by the author's company. But to me the key statement of the article is:
ReplyDelete"Today, sCMOS looks like it could become the next best thing for more generalized low-light imaging if one considers the combination of speed, dynamic range, and resolution as well as probable improvements in signal to noise on the horizon. And it will be interesting to see where EMCCDs fall once sCMOS finds its way into more applications."
I can't seem to find any reference to the MTF @ Nyquist provided by the sensor. anyone know what the MTF performance is? is it the same for vertical and horizontal directions?
ReplyDeletealso, how do they manage to get 16 bits out of two 11-bit ADC? There must be a heck of a dip/gap in the SNR vs. signal curve to stitch the two ranges together!