SWIR Camera

[ulf]

I do not remember seeing anything here about this camera:

https://trieye.tech/ovi/

 

I just found it in a newsletter for optical components and know nothing more about it.

[enricosavazzi]

Israel-based hi-tech companies like this one usually employ very sharp people. I know nothing about this specific company or product, but I suspect that, as usual, if you need to ask the price of this camera, you likely cannot afford it. Other than this, the camera is interesting because the 400-1,600 nm range bridges VIS, NIR, and SWIR (but does not extend into the multi-μm range where thermal IR gives interesting results with commonly available subjects). The sensor uses CMOS technology, so it is most likely silicon-based, rather than indium-gallium-arsenide-based. Ultimately, the intrinsic limitations of silicon place limits on the wavelengths that can be recorded by the sensor. Without knowing how the camera presents data of different wavelengths, and whether it can discriminate among different wavelengths, we cannot tell much more.

[Stefano]

Andy told us to keep three eyes on it: https://www.ultravioletphotography.com/content/index.php?/topic/4158-another-swir-company-to-keep-an-eye-on-or-three-of-them

[dabateman]

Yes just as I said in the previous thread,  expect it to mostly be a 1350nm swir camera: 

http://image-sensors-world.blogspot.com/2021/12/trieye-unveils-vcsel-powered-swir-camera.html?m=1

 

They do have a business partner now. So might get better for them:

http://image-sensors-world.blogspot.com/2022/01/swir-startup-trieye-collaborates-with.html?m=1

[dabateman]

This company maybe more interesting to watch:

https://seedevice.com/

 

From 300nm to 1600nm.

They just partnered with John Deere.

The engineering sample csn be seen here:

Engineering Sample Demo (PDF)

https://seedevice.com/wp-content/uploads/simple-file-list/210517-QPD-Visual-Eval-Board-300-1600nm-Product-Brief.pdf

[colinbm]

Thanks Dave

[Stefano]

From UVB to SWIR, that's quite a wide range. I would love such a camera.

[Guest]

Could you imagine walking into the grocery with one of those tri eye cameras going to the fruit/veggie section and scanning all your items for defects. I wonder if that would attract much attention. Actually, I could see that feature implemented in the iPhone 20 with many apps available for locating bad apples. Oh wait, my iPhone already identifies bad apples with its suite of cancel culture apps.

[Andy Perrin]

(Never mind, speculation deleted because I saw we already figured out what tech they are using in the previous thread on this.)

[Cadmium]

It looks cool. 

_{Does anyone see any lenses they have for it on there}?

We would need to know what lenses will work with it and the transmission.

Then you would need filters to define what range you want to capture, and the filters would need to be special for the higher range,

so I wonder if they have those special filters.
Other than the higher range, I suppose any lower range filter that we are familiar with should work for that?

[Andy Perrin]

Cadmium, most ordinary visible light lenses work adequately in SWIR with some focal shift. I have used achromatic doublets from Thorlabs also, which have a SWIR BBAR coating. Thorlabs also sells SWIR bandpass and longpass filters. I own a lot of them. 

[enricosavazzi]

On 1/27/2022 at 12:36 PM, dabateman said:

[...]

Engineering Sample Demo (PDF)

https://seedevice.com/wp-content/uploads/simple-file-list/210517-QPD-Visual-Eval-Board-300-1600nm-Product-Brief.pdf

"Photoelectric conversion efficiency 10,000%"

I wonder which galaxy they hail from?

[ulf]

Thermal cameras, if there is an unlimited budget, get two of these, just for fun:

https://www.allaboutcircuits.com/news/science-cameras-bring--thermal-imaging-to-pcb-and-hypersonic-research-alike/

 

“You can have one camera system targeting the 3–5 micron range and another camera system targeting the 7.5–12 micron range. We have the exact same interface for those cameras,” Lamont says. “Meanwhile, you'll be collecting different in-band radiance and temperature measurements.”

[Andy Perrin]

Isn’t defense research wonderful? All the money you want while the rest of your country makes due with what’s left over. 

[OlDoinyo]

I don't think the SeeDevice is a camera at all, but merely a brightness detector. 

[ultrainfra]

On 1/28/2022 at 2:40 AM, enricosavazzi said:

"Photoelectric conversion efficiency 10,000%"

I wonder which galaxy they hail from?

Might they be using the coma to indicate a decimal place instead of a period? I was under the impression most of the world uses "," as a decimal separator (including Sweden I thought?). 

[Andy Perrin]

2 minutes ago, ultrainfra said:

Might they be using the coma to indicate a decimal place instead of a period? I was under the impression most of the world uses "," as a decimal separator (including Sweden I thought?). 

One might think so, but it doesn't make a lot of sense that way either. A conversion efficiency of 10% would be pretty horrible.

[enricosavazzi]

6 hours ago, ultrainfra said:

Might they be using the coma to indicate a decimal place instead of a period? I was under the impression most of the world uses "," as a decimal separator (including Sweden I thought?). 

That is correct, but it would be an error to use a comma as decimal separator in an English text. A comma as decimal separator, followed by three zeroes, would also imply that they measured the % efficiency with a precision of five digits, which would be very difficult. A result with four significant zeroes not followed by other digits is highly suspicious in itself.

 

That said, a photon multiplier or a similar signal amplifier can realistically amplify a low-level signal by 10⁴ times. However, it is not correct to call this performance "photoelectric conversion efficiency" if the system involves an active amplification of the original signal. "Photoelectric conversion efficiency" is a term commonly used for photovoltaic panels, for example, which certainly do not do any active amplification because their job is to efficiently collect incident radiation and convert it to electrical energy, and obviously cannot exceed 100% conversion efficiency. And as Andy says, a 10% photon conversion efficiency would be pretty low for a current image/light sensor.

[Andy Perrin]

I would guess that whoever made the ad did not understand what they were writing and probably put a number in the wrong place. 

[Alaun]

9 hours ago, enricosavazzi said:

That is correct, but it would be an error to use a comma as decimal separator in an English text. A comma as decimal separator, followed by three zeroes, would also imply that they measured the % efficiency with a precision of five digits, which would be very difficult. A result with four significant zeroes not followed by other digits is highly suspicious in itself.

 

That said, a photon multiplier or a similar signal amplifier can realistically amplify a low-level signal by 10⁴ times. However, it is not correct to call this performance "photoelectric conversion efficiency" if the system involves an active amplification of the original signal. "Photoelectric conversion efficiency" is a term commonly used for photovoltaic panels, for example, which certainly do not do any active amplification because their job is to efficiently collect incident radiation and convert it to electrical energy, and obviously cannot exceed 100% conversion efficiency. And as Andy says, a 10% photon conversion efficiency would be pretty low for a current image/light sensor.

According to the patent description (EP 2 752 877 A2) they use -instead of a diode, collecting electrons- a kind of a transistor. The photons cause a tunneling and the generatd signal can be hundreds or more times higher than with a conventional sensor (before any additional signal treating is done).

[enricosavazzi]

17 hours ago, Alaun said:

According to the patent description (EP 2 752 877 A2) they use -instead of a diode, collecting electrons- a kind of a transistor. The photons cause a tunneling and the generatd signal can be hundreds or more times higher than with a conventional sensor (before any additional signal treating is done).

So it does use active amplification and therefore their use of the "photoelectric conversion efficiency" term is improper.

 

The "before any additional signal treating is done" is also partly incorrect/improper, because what a phototransistor does is actively amplify the received signal after converting photons to an electrical current, and therefore it does perform "additional signal treating".

[ultrainfra]

On 2/23/2022 at 2:56 AM, enricosavazzi said:

That is correct, but it would be an error to use a comma as decimal separator in an English text. A comma as decimal separator, followed by three zeroes, would also imply that they measured the % efficiency with a precision of five digits, which would be very difficult. A result with four significant zeroes not followed by other digits is highly suspicious in itself.

 

That said, a photon multiplier or a similar signal amplifier can realistically amplify a low-level signal by 10⁴ times. However, it is not correct to call this performance "photoelectric conversion efficiency" if the system involves an active amplification of the original signal. "Photoelectric conversion efficiency" is a term commonly used for photovoltaic panels, for example, which certainly do not do any active amplification because their job is to efficiently collect incident radiation and convert it to electrical energy, and obviously cannot exceed 100% conversion efficiency. And as Andy says, a 10% photon conversion efficiency would be pretty low for a current image/light sensor.

Thanks for clarifying! 

I hope SWIR imaging tools become available in an affordable form someday. Of course the demand for such devices is probably low enough that prices will remain high. 

[Daniel Csati]

There are sometimes SWIR line scan cameras for sale on eBay. I bought recently for $850 this 1024pixel InGaAs camera.

https://www.ebay.com/itm/294343549549

For snapshots it will do fine, just needs some rotation mechanism and a portable framegrabber

 

[Stefano]

Looking forward to see what you get from it. It should work well if you scan at a constant speed.

[Daniel Csati]

I did it earlier with a VIS-NIR pushbroom hyperspectral imager. It should be easy to set this up, I just hope that the camera works fine..

The concept is following: camera and frame grabber is powered from a AC powerbank, for rotation I use a Syrp Genie Mini and the laptop to record the images. Disadvantage of the setup is that I can only record completely horizontal images and it's a heavy kit.

Camera should arrive soon and I'm looking forward to playing with it!