Author Topic: Mµ Thermal Imager - real or fake?  (Read 377656 times)

0 Members and 1 Guest are viewing this topic.

Offline mikeselectricstuff

  • Super Contributor
  • ***
  • Posts: 11942
  • Country: gb
    • Mike's Electric Stuff
Re: Mµ Thermal Imager - real or fake?
« Reply #1175 on: September 14, 2014, 12:44:45 pm »
That was the IRISYS people counter that Mike cleverly re-purposed with a custom interface so that it was capable of far more than the OEM intended. IRISYS are a UK company that decided to make its own thermal detector FPA's to meet its needs.

Oh! I thought it was completely Mike's work, but I stand corrected.

They released thermal cameras that used a IPAQ PPC as its display and processing. The People counter used IIRC a 16 x 16 matrix pyroelectrric detector array. It was able to track movement within its field of view and analyse the target as valid or invalid.

I was actually thinking about getting one of these (either a people counter or the iPaq one), as they go for a little over free on eBay, but the resolution is a joke for a thermal cam...
Don't - the camera ones are total rubbish.
The people counters aren't really useable as cameras as they have no shutter so only respond to changes - any static image will fade out over 5-10 seconds. They work brilliantly at what they are designed for, but not much use for anything else.
Youtube channel:Taking wierd stuff apart. Very apart.
Mike's Electric Stuff: High voltage, vintage electronics etc.
Day Job: Mostly LEDs
 

Online Fraser

  • Super Contributor
  • ***
  • Posts: 8321
  • Country: gb
Re: Mµ Thermal Imager - real or fake?
« Reply #1176 on: September 14, 2014, 02:37:17 pm »
+1 on Mike's comments about the IRISYS units.

In the UK the IRISYS IPAQ based cameras sell for well over GBP300 each and are not value for money as the low resolution just creates areas of colour on an image rather than the outline of the heat producing source. This makes it hard to identify exactly where the heat energy is centred. Totally useless for PCB work. If you want to see the thermal resolution in action, take a look at the FLUKE VT02 without the visible image. Dave did a test of the VT02 and even with the visible image you will see the low thermal resolution.

As Mike stated the people counters are without a shutter wheel as the designer wanted the end product to only see moving targets. This meant that static heat sources like a power supply or radiator would be automatically ignored. I spoke with IRISYS about the modification of a people counter into a camera..... the engineer didn't think it was worth the effort. Mike now owns my unit as he has better uses for it than me  :)
« Last Edit: September 14, 2014, 02:43:10 pm by Aurora »
 

Offline callipso

  • Regular Contributor
  • *
  • Posts: 87
  • Country: cz
Re: Mµ Thermal Imager - real or fake?
« Reply #1177 on: September 14, 2014, 06:11:44 pm »
That was the IRISYS people counter that Mike cleverly re-purposed with a custom interface so that it was capable of far more than the OEM intended. IRISYS are a UK company that decided to make its own thermal detector FPA's to meet its needs.

Oh! I thought it was completely Mike's work, but I stand corrected.

They released thermal cameras that used a IPAQ PPC as its display and processing. The People counter used IIRC a 16 x 16 matrix pyroelectrric detector array. It was able to track movement within its field of view and analyse the target as valid or invalid.

I was actually thinking about getting one of these (either a people counter or the iPaq one), as they go for a little over free on eBay, but the resolution is a joke for a thermal cam...
Don't - the camera ones are total rubbish.
The people counters aren't really useable as cameras as they have no shutter so only respond to changes - any static image will fade out over 5-10 seconds. They work brilliantly at what they are designed for, but not much use for anything else.

Since HeMa flopped I'll probably wait for a Seek or HeMa relaunch..
Also anyone selling a TIC?
I wonder... **BOOM
 

Offline CanadianAvenger

  • Regular Contributor
  • *
  • Posts: 178
Re: Mµ Thermal Imager - real or fake?
« Reply #1178 on: September 15, 2014, 09:17:27 pm »
And we have pictures!!

Quote
Hello everyone. 

Saturday evening marked the end of the International Machine Tool Show here in Chicago.  While it might not seem to be the most natural place to debut the camera; we decided to show there as a developmental partner to GTI Spindle; who were already showing their wireless Predictive Maintenance hardware/software.   We did not put out any press releases about our attendance because it wasn't our
place to do that, we were the guests of another company.  They were a generous host and we are extremely thankful for their support. 

At the end of the show, I believe I speak for everyone involved when I say that the convention was a solid showing for the camera.  Everyone who saw it was a bit in awe of it's capabilities, simplicity of operation, overall quality, and price. 

I've attached a few more simple images for your perusal.

Thanks to everyone for your support and especially to those who stopped by the booth at the show and expressed such enthusiasm for the camera they saw. 

More news to follow in the next couple of weeks.

Thanks,

John McGrath

 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 29260
  • Country: au
    • EEVblog
Re: Mµ Thermal Imager - real or fake?
« Reply #1179 on: September 15, 2014, 10:34:55 pm »
And we have pictures!!

So how long did that take since they said showing a thermal image was their #1 priority?  :palm:
 

Offline CanadianAvenger

  • Regular Contributor
  • *
  • Posts: 178
Re: Mµ Thermal Imager - real or fake?
« Reply #1180 on: September 16, 2014, 02:30:37 am »
And we have pictures!!

So how long did that take since they said showing a thermal image was their #1 priority?  :palm:

No kidding!
 

Offline CanadianAvenger

  • Regular Contributor
  • *
  • Posts: 178
Re: Mµ Thermal Imager - real or fake?
« Reply #1181 on: September 16, 2014, 10:03:40 am »
Oh, and since they are static app screen grabs, there is no way to actually confirm that the images are from their sensor, and not ones imported from a commercial unit... so still not REAL proof of a functioning prototype. The best proof we have so far is from KentuckyEE! By these Mu Optics guys really suck at this!!
 

Online dunkemhigh

  • Super Contributor
  • ***
  • Posts: 1381
Re: Mµ Thermal Imager - real or fake?
« Reply #1182 on: September 16, 2014, 10:18:45 am »
It has been playing on my mind somewhat for a while that Mu might actually pull this off. Maybe not strictly to the letter of their original spec/cost, but very different from the walking (or running!) away that most people here have stated will be a fact. I wonder how we will react if that happens?

Already, Dave has downgraded from an absolute 'never going to happen' to '#1 priority took a long time'. Of course, it may still never happen, but maybe we need to consider appropriate fallback (and face saving) positions just in case :)
 

Offline callipso

  • Regular Contributor
  • *
  • Posts: 87
  • Country: cz
Re: Mµ Thermal Imager - real or fake?
« Reply #1183 on: September 16, 2014, 11:23:39 am »
Quote
And we have pictures!!

160*120 my arse, the image is square and interpolated!

Also the hairdryer looks quite cold, my girlfriend's hairdryer nozzle gets rather hot, at least hotter than body temperature, but in the picture it looks about that...
I wonder... **BOOM
 

Online dunkemhigh

  • Super Contributor
  • ***
  • Posts: 1381
Re: Mµ Thermal Imager - real or fake?
« Reply #1184 on: September 16, 2014, 12:13:22 pm »
Quote
girlfriend's hairdryer nozzle gets rather hot

Don't forget that the Mu measurement range is severely limited (the primary reason I dropped out and got a refund early on). Can't recall the actual figure, but I think body temperature is very close indeed to the top end.
 

Online rs20

  • Super Contributor
  • ***
  • Posts: 2142
  • Country: au
Re: Mµ Thermal Imager - real or fake?
« Reply #1185 on: September 16, 2014, 01:05:29 pm »
Quote
girlfriend's hairdryer nozzle gets rather hot

Don't forget that the Mu measurement range is severely limited (the primary reason I dropped out and got a refund early on). Can't recall the actual figure, but I think body temperature is very close indeed to the top end.

So the nozzle should be saturated off-scale high, right?
 

Offline callipso

  • Regular Contributor
  • *
  • Posts: 87
  • Country: cz
Re: Mµ Thermal Imager - real or fake?
« Reply #1186 on: September 16, 2014, 01:09:07 pm »
Quote
girlfriend's hairdryer nozzle gets rather hot

Don't forget that the Mu measurement range is severely limited (the primary reason I dropped out and got a refund early on). Can't recall the actual figure, but I think body temperature is very close indeed to the top end.

Yes, but the hot patch on his/her head is visibly significantly hotter than the rest of the body, whereas the hairdryer is about as hot as the body. Seems odd to me.

So the nozzle should be saturated off-scale high, right?

That ^
I wonder... **BOOM
 

Online dunkemhigh

  • Super Contributor
  • ***
  • Posts: 1381
Re: Mµ Thermal Imager - real or fake?
« Reply #1187 on: September 16, 2014, 02:06:46 pm »
Quote
So the nozzle should be saturated off-scale high, right?

I would expect any over-temperature to display at the highest value.

Looking again at the photo now, I am not convinced it is a fake. We are seeing the temperature of the nozzle, not the air or the element. Depending on design the nozzle could be quite cool, and it might be quite reflective. I don't actually see anything wrong with that part of the photo - things are not always as you think they should be, and that's when you can see the original for real.
 

Offline mikeselectricstuff

  • Super Contributor
  • ***
  • Posts: 11942
  • Country: gb
    • Mike's Electric Stuff
Re: Mµ Thermal Imager - real or fake?
« Reply #1188 on: September 16, 2014, 02:21:10 pm »
Quote
So the nozzle should be saturated off-scale high, right?

I would expect any over-temperature to display at the highest value.

Looking again at the photo now, I am not convinced it is a fake. We are seeing the temperature of the nozzle, not the air or the element. Depending on design the nozzle could be quite cool, and it might be quite reflective. I don't actually see anything wrong with that part of the photo - things are not always as you think they should be, and that's when you can see the original for real.
After the show I don't think there is any doubt they have something that appears to work.
However that is still potentially a very long way from delivering the the 320x240 30fps imager for $300 they were promising.
Still filed under "nothing to see here" for the forseeable future.
Youtube channel:Taking wierd stuff apart. Very apart.
Mike's Electric Stuff: High voltage, vintage electronics etc.
Day Job: Mostly LEDs
 

Offline Rasz

  • Super Contributor
  • ***
  • Posts: 2245
  • Country: 00
    • My random blog.
Re: Mµ Thermal Imager - real or fake?
« Reply #1189 on: September 16, 2014, 02:34:52 pm »
Now that they appear to have working prototype, they should launch a Kickstarter campaign with delivery date somewhere 9-12 months in the future. I think they should be ready to ship on about 2 years.
Who logs in to gdm? Not I, said the duck.
My fireplace is on fire, but in all the wrong places.
 

Offline Jay_Diddy_B

  • Super Contributor
  • ***
  • Posts: 1788
  • Country: ca
Re: Mµ Thermal Imager - real or fake?
« Reply #1190 on: September 16, 2014, 03:33:52 pm »
Hi group,

Here is a comparison of the cup of tea image;

Mu Optics




Fluke Ti9

Image data: Royal Doulton Lambeth ware cup and saucer with hot water and a stainless steel teaspoon.

Fluke Ti9 Thermal Image 160 x 120

Same image shown with different pallets.









Regards,

Jay_Diddy_B

« Last Edit: September 16, 2014, 03:35:26 pm by Jay_Diddy_B »
 

Online dunkemhigh

  • Super Contributor
  • ***
  • Posts: 1381
Re: Mµ Thermal Imager - real or fake?
« Reply #1191 on: September 16, 2014, 03:34:30 pm »
Quote
still potentially a very long way from delivering

Indeed. But that's infinitely closer than not having anything at all :)
 

Offline tom66

  • Super Contributor
  • ***
  • Posts: 3414
  • Country: gb
  • Electronic Engineer & Hobbyist
Re: Mµ Thermal Imager - real or fake?
« Reply #1192 on: September 16, 2014, 03:41:41 pm »
The resolution is much lower than 160x120.

Image is 360 pixels across, some noise pixels are about 4~6 pixels wide.

Best guess, 80 or 60 pixels wide sensor. (Could be same sensor as that 60x62 thermal camera that's actually not vaporware.)
 

Offline callipso

  • Regular Contributor
  • *
  • Posts: 87
  • Country: cz
Re: Mµ Thermal Imager - real or fake?
« Reply #1193 on: September 16, 2014, 05:48:28 pm »
Quote
So the nozzle should be saturated off-scale high, right?

I would expect any over-temperature to display at the highest value.

Looking again at the photo now, I am not convinced it is a fake. We are seeing the temperature of the nozzle, not the air or the element. Depending on design the nozzle could be quite cool, and it might be quite reflective. I don't actually see anything wrong with that part of the photo - things are not always as you think they should be, and that's when you can see the original for real.
After the show I don't think there is any doubt they have something that appears to work.
However that is still potentially a very long way from delivering the the 320x240 30fps imager for $300 they were promising.
Still filed under "nothing to see here" for the forseeable future.

Owing to the lag I think it might just be playing back a doctored video, I mean... they are video guys who had more than plenty of time.
OR it might be that ref board they supposedly had in the last photo-update.
« Last Edit: September 16, 2014, 05:55:02 pm by callipso »
I wonder... **BOOM
 

Offline callipso

  • Regular Contributor
  • *
  • Posts: 87
  • Country: cz
Re: Mµ Thermal Imager - real or fake?
« Reply #1194 on: September 16, 2014, 05:59:55 pm »
OR it might be a Heimann sensor, the one HeMa uses - resolution just about fits and the sensor is in fact noisy, like the pictures. Hook that up to a little Arduino, toss in a wlan module for 6$ off of eBay (the sluggish speed) and you have a ... $1000 TIC (the Heimanns are this expensive in low OQ). That's amount of money they certainly have.

Why go this far with a scam product? They're now in no way liable if they say they're out of cash. They tried, showed a prototype and failed. Pocket the remainder and there you go.
I wonder... **BOOM
 

Offline Simon P

  • Newbie
  • Posts: 4
  • Country: au
Re: Mµ Thermal Imager - real or fake?
« Reply #1195 on: September 17, 2014, 08:53:24 am »
Quote
So the nozzle should be saturated off-scale high, right?

I would expect any over-temperature to display at the highest value.

Looking again at the photo now, I am not convinced it is a fake. We are seeing the temperature of the nozzle, not the air or the element. Depending on design the nozzle could be quite cool, and it might be quite reflective. I don't actually see anything wrong with that part of the photo - things are not always as you think they should be, and that's when you can see the original for real.
After the show I don't think there is any doubt they have something that appears to work.
However that is still potentially a very long way from delivering the the 320x240 30fps imager for $300 they were promising.
Still filed under "nothing to see here" for the forseeable future.

Mike,

I am confused.  When you say "a very long way from delivering the the 320x240 30fps imager " are you still talking about Mu Optics ?  I have been following this IDGG campaign with great scepticism and and been amused by all the various comments by the backers on IDGG since it was first mentioned on this forum .  It's a great source of entertainment.  But even with all their various BS, I still don't ever recall them quoting 320x240.  In the original story they say 160 x 120 ?

The first thing I actually noticed was the really poor quality case on their "prototype".  I reckon that someone has knoocked the case up on a 3D printer.  If it is indeed injection molded (Doubtful), then I would also wonder about the quality of what you CANNOT actually see on the inside or how long it would even last !   See attached screen grab.


Simon
Simon
 

Offline mikeselectricstuff

  • Super Contributor
  • ***
  • Posts: 11942
  • Country: gb
    • Mike's Electric Stuff
Re: Mµ Thermal Imager - real or fake?
« Reply #1196 on: September 17, 2014, 09:06:09 am »
Quote
So the nozzle should be saturated off-scale high, right?

I would expect any over-temperature to display at the highest value.

Looking again at the photo now, I am not convinced it is a fake. We are seeing the temperature of the nozzle, not the air or the element. Depending on design the nozzle could be quite cool, and it might be quite reflective. I don't actually see anything wrong with that part of the photo - things are not always as you think they should be, and that's when you can see the original for real.
After the show I don't think there is any doubt they have something that appears to work.
However that is still potentially a very long way from delivering the the 320x240 30fps imager for $300 they were promising.
Still filed under "nothing to see here" for the forseeable future.

Mike,

I am confused.  When you say "a very long way from delivering the the 320x240 30fps imager " are you still talking about Mu Optics ?  I have been following this IDGG campaign with great scepticism and and been amused by all the various comments by the backers on IDGG since it was first mentioned on this forum .  It's a great source of entertainment.  But even with all their various BS, I still don't ever recall them quoting 320x240.  In the original story they say 160 x 120 ?

Maybe I was mis-remembering what they were claiming through all the BS - same argument applies to 160x120 though. I don't believe any sensor+lens currently exists that can be incorporated in a unit selling for $300

 
Youtube channel:Taking wierd stuff apart. Very apart.
Mike's Electric Stuff: High voltage, vintage electronics etc.
Day Job: Mostly LEDs
 

Offline eneuro

  • Super Contributor
  • ***
  • Posts: 1411
  • Country: 00
Re: Mµ Thermal Imager - real or fake?
« Reply #1197 on: September 17, 2014, 09:54:12 am »
very long way from delivering the the 320x240 30fps imager for $300
Do we really need 30fps in those things?
Even at 1fps at some applications when we make 10 shots in the 10s time and average those images we can get quite accurate energy leakage view while looking at stationary objects.
Postprocessing of this data and correct analysis is the most important there.
In my opinion such device should simply have wifi module, everything hermeticaly closed and be able send its imaginery at least 10meters away for a few hours, so it could be easy captured, stored, postprocessed and analysed not only on those crappy "smartphones" with small screens but everything that has wireless network.
Plugging together this Mu to smartphone is hopeless  :palm:
There are situations where you want be far away from heat source or its very cold outside and I want monitor my window in another room sitting comfortable and playing with those images on 24" iiyama screen using OpenCV for autodetection of interesting areas.
12oV4dWZCAia7vXBzQzBF9wAt1U3JWZkpk
“Let the future tell the truth, and evaluate each one according to his work and accomplishments. The present is theirs; the future, for which I have really worked, is mine”  - Nikola Tesla
-||-|-
 

Offline mikeselectricstuff

  • Super Contributor
  • ***
  • Posts: 11942
  • Country: gb
    • Mike's Electric Stuff
Re: Mµ Thermal Imager - real or fake?
« Reply #1198 on: September 17, 2014, 10:46:21 am »
very long way from delivering the the 320x240 30fps imager for $300
Do we really need 30fps in those things?

No we don't and the export regs issue means it would be even more of a pain to manufacture and sell.
I have no doubt that mu were completely unaware of this issue when they started - pretty sure I saw 30fps mentioned somewhere. I don't know if they've ever said anything about it subsequently. 30fps would probably also be hard to sustain over a wireless connection, especially bluetooth
 
Youtube channel:Taking wierd stuff apart. Very apart.
Mike's Electric Stuff: High voltage, vintage electronics etc.
Day Job: Mostly LEDs
 

Offline eneuro

  • Super Contributor
  • ***
  • Posts: 1411
  • Country: 00
Re: Mµ Thermal Imager - real or fake?
« Reply #1199 on: September 17, 2014, 11:33:04 am »
30fps would probably also be hard to sustain over a wireless connection, especially bluetooth
320x240 is 76800 pixels -2 bytes per pixel (16bits)  gives 150KB, so at 30fps about 4.4MB/s if we want raw data, while any non loosing compression attempts on this thermal imager device will require extra complexity and compression processing power losses unless we have chip cappable of doing such things implemented and accelerated in hardware.
12oV4dWZCAia7vXBzQzBF9wAt1U3JWZkpk
“Let the future tell the truth, and evaluate each one according to his work and accomplishments. The present is theirs; the future, for which I have really worked, is mine”  - Nikola Tesla
-||-|-
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf