Friday, September 10, 2021

Smartsens Upgrades Two Sensors for Security and Automotive Applications

SmartSens releases the 1/3-inch, 2MP SC2336 and 1/2.8-inch, 3MP SC3336—two performance upgrade CMOS sensors with DSI-2 technology, aiming at surveillance and automotive camera applications. Benefiting from PixGain technology, the SC3336 supports High Gain and Low Gain to adjust to light changes between day and night.

Both SC2336 and SC3336 are equipped with an ultra-low-noise readout, combined with cinema-level color visual effects technology for more detailed and more realistic night vision full-color imaging. Taking SC2336's Chroma as an example, the low color temperature increased by 15% , and the high color temperature increased by 11.8%.

In addition, the electrical crosstalk (Blooming) of SC2336 and SC3336 is significantly reduced by 45% and 35% , respectively , compared with the previous generation products, which improves the flashing that is easily generated by the camera in backlit scenes and night shooting lights.

The SC2336 has achieved mass production and the SC3336 is available for sampling now, with mass production expected in Q4 of 2021.

14 comments:

  1. Anybody knows what the DSI means compared to FSI and BSI? Is it standing for double side illumination or diffuse surface illumination?

    ReplyDelete
    Replies
    1. My understanding is that "D" in DSI stands for Dongbu Hitek foundry that manufacturers this sensor. This is somewhat similar to Galaxycore's TSI branding for the sensors manufactured by TSMC. Other than the branding, these are supposed to be BSI sensors.

      Delete
    2. Thank you for your reply Vladimir. Since in their ad, it claims the QE is 2x higher and dark current is 5x lower compared to BSI, then I thought it might be something new.

      Delete
    3. Albert Theuwissen - Harvest ImagingSeptember 11, 2021 at 1:17 PM

      Be carefull !!! QE is not the same as sensitivity in mV/lux.sec !!! As far as I can judge (I do not understand Chinese), the light source is not specified and in that case you can "put" into a lux whatever you want. Do you realize that an imager can have x mV/lux.sec output for 0 lux input ?

      Delete
    4. This comment has been removed by the author.

      Delete
    5. My argument has nothing to do with any company or country, my statement has to do with technical correctness. I will never publicly argument against my own customers !

      Delete
    6. The Unknown's statement is very strange. I would guess you are from Smartsens and you got emotional. I don't think Albert had any implication against you (may be Smartsens) or China, and probably you don't know Albert personally. The fact is that Albert had taught so many Chinese students (probably even majority of his PhD/Master students were from China, they are now contributing a lot for either Chinese companies or companies in Western countries.

      Albert's statement is purely based on technical correctness. Be fair, if you don't mention your wavelength or lighting conditions (like use D65 light) when you talking about mV/lux s, it is meaningless, and so misleading. Without mentioning lighting conditions, I can make a sensor with sensitivity 100 times higher than Smartsens's sensor.

      Delete
    7. For the Unknown person, this is a place for technical discussion, not a place for your wolf warrior spirit. If I am an engineer from OV, I would feel shame seeing this comment, as like got bought by Chinese money is to "teach" everyone else a lesson. Disclaimer: I am also a chinese researcher, also want to make contribution to the technology as a chinese, but not in the way of "teaching"

      Delete
    8. I also fully support the comments from previous 2 'Anonymous'. I am also a Chinese, and feel very uncomfortable for what 'Unknown' said to Albert.

      For the 'Unknown', please tell me which word Albert used made you think that Albert thought Chinese companies cannot make the best product. And it is so wrong that you said you want to 'teach' Albert. When Albert could design the best sensors in the world, probably you were not even born. Even that was not Albert, only illiterate person would attack people by saying 'teach you'.

      Delete
    9. I come from China, I feel so shame and uncomfortable for the "Unknow" person, I can not image there exists unfriendly person even in the pure academic society. I guess everyone who learn image sensor knows and respects Prof. Albert!

      Delete
    10. I never really understand why people still using V/Lux*s, it is non-sense spec, not only about lux, but also Volt? is it pixel output? PGA output? it is not measureable and not accurately definied. Feel strange actually why there are still (big) companies stick to this spec.
      As for the "biased"comments, Prof Theuwissen is my Ph.D Supervier, a tutor/mentor in my professional and personal life and a dear friend of mine. And I am a Chinese, managing a chinese image sensor company. None of this would have happened without Albert. I am pretty sure he will not be "biased" based on nationality, etc

      Delete
  2. Yes, thank you for the correction, not QE, sensitivity can be improved by other ways.

    ReplyDelete
  3. Also hopefully it is not DC when the input is 0 lux.

    ReplyDelete
    Replies
    1. Albert Theuwissen - Harvest ImagingSeptember 12, 2021 at 1:47 PM

      Any light input with a spectrum OUTSIDE the visible band is by definition 0 lux. So if you illuminate a sensor with an LED that has a wavelength of 850 nm, independent of the amount of light energy emitted by the LED, it is always 0 lux.
      BTW, talking about DC or dark current, check out this one :
      https://www.harvestimaging.com/forum_introduction_2021_new.php

      Delete

All comments are moderated to avoid spam and personal attacks.