Thermal Imaging and IPhone machine vision... / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Thermal Imaging and IPhone machine vision...

1 2 next

Move Thread

LAN_403

Patrick

Sun Aug 24 2014, 05:34AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

ive been looking here :

It looks like :

Im thinking of intercepting the normal iphone connector and using this in drones and embeded aplications. The problem is i dont want to use or have a Iphone 5. Im thinking i should contact them as a college and see if theyll let us have an experimental version for academic use.

If i have a second cam which blocks red, but sees IR+B+G then i can really have a sensor suite.

Bjørn

Sun Aug 24 2014, 10:59AM

Registered Member #27 Joined: Fri Feb 03 2006, 02:20AM
Location: Hyperborea
Posts: 2058

If you are really lucky it works as a webcamera, their stand alone units works as web cameras when connected to a PC.

klugesmith

Mon Aug 25 2014, 06:37AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1714

Thanks for sharing, Patrick.

I took issue with FLIR's statement that "Most of the energy of the sun arrives on earth as infrared light." Not because 99.999... percent of the energy misses the earth.

But I thought our visible spectrum is so well matched to our sun's color that it includes most of the total irradiance.

Google found me this document: "Solar constant versus the electromagnetic spectrum"

which figures the solar constant to be 1340 W/m^2, spectrally distributed as a blackbody of 5776 K. They integrate the power spectral density and get (in Table 1):
12% UV (shorter than 400 nm)
37% visible (400 to 700 nm)
51% IR (longer than 700 nm).

OK, so I was wrong. But Agrawal's wavelength boundaries give too much to the UV and IR. CIE standard observer functions are tabulated from 380 to 780 nm, and I can attest that 850 nm datacomm lasers are visible if you look into them.

This link cites references pegging the boundary at 700, 720, 750, 770, and 780.

So I think there's a strong case that the FLIR statement is untrue, barely. At least if we consider the Earth to start at the top of our atmosphere.

Steve Conner

Mon Aug 25 2014, 06:58AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Like all good advertising copy, the statement is untrue on several levels and irrelevant on several more.

Firstly, long-wave infrared of the sort picked up by thermal imaging cameras is blocked by the so-called "greenhouse gases" in the Earth's atmosphere (unless you're a Republican) There is still long-wave infrared in ambient light, but most of it doesn't come directly from the sun.

Secondly, the vast majority of applications for thermal cameras involve looking at man-made (or other animal) heat sources. IR from the sun is just a nuisance that forces you to use the camera at night. I can only think of one that doesn't, and that is detecting gas leaks with a narrowband optical filter. With just the right filter, the gas looks like dark smoke against the ambient IR background.

Carbon_Rod

Mon Aug 25 2014, 10:00AM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

You may want to cite somebody other than Agrawal...
Perhaps the credulous generate animosity toward critical thought, but Hanlon's razor may explain why skewed spectrum distribution and energy emission peaks are a difficult subject for some.

IIRC, early CCD camera modifications with the TEC stack could allow them to operate well into the deep IR...
Modern elements still calibrate the sensor instead to avoid the sensor die temperature control problem.
Glass optics will unlikely work very well, but a gold catadioptric optical system may...
...if you 3D printed a mount that works it may prove an interesting project...

Patrick

Mon Aug 25 2014, 05:18PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Hanlon's Razor ? Well that explains Nancy Pelosi !

I'm just wanting to get decent thermal views from the air, to a memory card.

I'm wanting: (all "shuttered" via micro controller simultaneously. )
- thermal cam pics
- NIR+B+G cam pics
- RGB cam pics

This way, after flight, I'll be able to get Normalize Difference Vegetation Index, (NVDI). then I'll make flights for farmers. Right now in California we are in extreme and exceptional drought, the worst in 20+ years supposedly. So maybe I could make money.

and a live VGA down link to my helmet monitor for real time flight.

so I'm thinking I'll need at least 2 mobius cams, which are about 70 US $ each.
I'm still trying to figure out which zigbee/xbee is right for this purpose. (Shutter trigger, not pic transmission to ground)

klugesmith

Mon Aug 25 2014, 08:50PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1714

Patrick wrote ...
... So maybe I could make money. ...

As I'm sure you know, the U.S. FAA is being sued by drone interests, to relax their long-standing ban on commercial use of unmanned aerial vehicles. Here's one link that presents the FAA side of the argument.

Patrick

Mon Aug 25 2014, 09:26PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

klugesmith wrote ...

Patrick wrote ...
... So maybe I could make money. ...

Reading these FAA talking points is like reading anti-western CCCP garbage.
You see how only two commercial operations have been approved, both above the arctic, both from big makers. That's not coincidence, similar trickery was used in the "Harrison stamp tax act", the FAA wants the big makers like Boeing, Lockheed to right the rules from K-Street.

the real mistake the FAA made was the 6 random sites, they offered which are useless, here in the states.

Yep, we're sticking together on these matters at the RCGroups forum. The FAA likes to huff and puff, but they also know that we commercial operators are many. And that we band together, while looking for a FAA action that gives one of us, (Supported by all others) the legal standing to take them all the way up to the US supreme court, as they're previous decisions (and the FAA knows it) are legally flawed. Many in my home town have already been making money for the past year. So, there are others bigger than myself.

The truth is the FAA is in over their head, and at least they're smart enough not to use heavy handed tactics ... yet. That will really unify, us. And potentially case wide spread civil disobeidence from flyers of all types, which would prove the FAA incompetent.

Steve Conner

Tue Aug 26 2014, 06:46AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Staying on topic... Seems like the easiest way to trigger the shutter would be a spare channel on your RC transmitter? Or program the GPS to take a picture every so many meters, giving you a grid of images that you can stitch together into one huge one.

Patrick

Tue Aug 26 2014, 07:10AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Steve Conner wrote ...

Staying on topic... Seems like the easiest way to trigger the shutter would be a spare channel on your RC transmitter? Or program the GPS to take a picture every so many meters, giving you a grid of images that you can stitch together into one huge one.

on the mobius cam im sure i can wire the shutter buttons and use an Xbee/Zigbee to trigger them for now... maybe GPS in a more advanced version, but the inertial measurement unit may be better for stitching pics into maps.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob