Forum - CMOS Sensor Choice

Hi Greg,

Great question. In general one would want a CMOS sensor with the following:

1. Small footprint and low weight
2. large pixel size
3. High sensitivity
4. Resolution high enough for imaging cell bodies

The MT9V032 nicely fulfills points 2 and 4 but there are definitely better options on the market. Our main reason for initially picking the MT9V032 is it supported an LVDS output capable of sending imaging data over a meter or two of cable. We used this LVDS in version 1 of our scopes but were not happy with its stability and performance (from my understanding Inscopix scopes send data over LVDS pairs and have issues with stability as well). In version 2 of the scopes (which is what is available on this site) we completely redesigned our electronics to serialize the data off the CMOS chip and communicate over a single coax cable instead of the twist wire pairs needed for LVDS. The changes made in version 2 mean we no longer rely on needing a CMOS sensor with LVDS output and instead can support the vast majority of commercial sensors on the market. At some point we would like to switch from the MT9V032 to a better and small sensor but this is somewhat on the back burner. While there is a bit of weight and sensitivity to be gained by switching over to a new sensor, these gains are pretty small.

Just a few other points to be made about CMOS sensors in general:

• Larger pixel size generally will result in better signal to noise. Since we have size and weight restrictions for the sensor in our system, having large pixels will also limit our options for resolution. In my opinion, once you reach ~0.2M pixels there is not much to be gained from higher resolution. I would rather have larger pixels and lower resolution than vice versa.
• E2V has a nice CMOS sensor which would be an overall upgrade to the MT9V032. ~0.5M pixels, smaller footprint, lower power consumption, and higher sensitivity.

Thank you! This is very helpful.