Flir E4 Thermal imaging camera teardown

[London Lad]

May not be 'nice' but hardly fraud.

[Fraser]

Wow ! $1000 profit already.

Ref. selling upgraded products for profit.

This is not uncommon. People do not realise how straight forward it is to enhance the camera and feel they are getting a bargain when someone else has done the work for them.

It is a bit entrepreneurial but not really 'wrong' in fraud terms as the seller has made it VERY clear that this is an E4 that has been enhanced to perform LIKE and E8. At no point does he claim it is an E8 or identical to such. Pretty honest but as you say, it would have been nice to credit the authors of the enhancement or at least the forum.

I suppose if someone wishes to effectively pay $1000 for the time a person spent doing the enhancement, that is their business.

As to FLIR, I doubt that they will be that concerned provided people do not try to pass off an E4 as a genuine E8. Such would be fraud and they would likely report the listing as their only action. If it became a mass sale of such fraudulently described devices, they might take legal advice on stopping such.

[Fraser]

For info only....

Junktronix has kindly sent me some high quality Thorlabs Germanium lenses to experiment with.

I shall see if they have an application with the E4. These lenses will be superior in performance to ZnSe if their characteristics permit their use. They appear to be suitable as a primary lens with a back focus distance of around 15mm

http://www.ebay.co.uk/itm/111231059741?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1423.l2649

Datasheet

http://www.thorlabs.de/newgrouppage9.cfm?objectgroup_id=1780&pn=LA9410-F

I will advise on their usefulness to us as soon as I have done some tests (hopefully this week)

[m4rkiz]

for all those that are not sure if they want to install menu hack






and for those that didn't decided if they want to get additional lens

[Fraser]

Thanks for the videos.

It is impressive how far the humble budget E4 has come thanks to the efforts of some generous members of this forum who shared their knowledge with us all.  I have said it before, but thanks to all who have helped to develop the E4 baseline in to what it is today....an E8+ 

[Bonocr]

@m4rkiz

Thanks for the video, very useful.
I just bought the lens

[Bonocr]

@freak_ge

Unfortunately, you're right 

[digsys]

It is a bit entrepreneurial but not really 'wrong' in fraud terms as the seller has made it VERY clear that this is an E4 that has been enhanced to perform LIKE and E8. At no point does he claim it is an E8 or identical to such. Pretty honest but as you say, it would have been nice to credit the authors of the enhancement or at least the forum.   

Has he explained that IF it ever goes in for repair, or gets a firmware update, the "extras" are GONE !!

It's not like anyone will ever repair these themselves, and this not an unlikely scenario. IF he'd pointed to this thread,
THEN said "I'm charging the extra for my work", fine, but he hasn't pointed out the pitfalls !! AFAIC it stinks.

[steve1515]

Is it possible to make the fine adjust distance menu display in feet instead of meters?

[Fraser]

@digsys

Two Words....... CAVEAT EMPTOR

When spending this sort of money, the bidder would be foolish to not ask questions, consider implications and research the product 

It is, after all, an open auction and not a classified ad. Bidders pay what they think an item is worth to them.

As long as the auction text is factually accurate, it is not the job of the seller to protect bidders from themselves 

[Fraser]

For those, like me, who are amazed at the price that an E4-8 is achieving on eb*y.....think what will happen if/when FLIR make the improvement harder or not feasible ..... enhanced units or those that can be enhanced will become even more desirable. 100% profit will then be the norm ?

That's why I ordered 10 E4's and stored them in my garage 

Just kidding 

[Rufus]

When spending this sort of money, the bidder would be foolish to not ask questions, consider implications and research the product 

There are millions of buyers on ebay - foolish ones are easily found.

[Fraser]

There is nothing to suggest that they are or are not taking action on this matter already.

It is not unreasonable to believe that FLIR are taking the matter seriously as  it does effect their PR side if it becomes too well known. To develop a modified firmware and thoroughly test it takes time and I will not be at all surprised if some level of countermeasure is contained in a new firmware for release in the future. That is just the way these things naturally go. Rigol is a perfect example of such attempts at countermeasures when the DS1052 was improved. Not very successful countermeasures in that case though. 

FLIR are a very powerful and professional company in the world of thermography....if they really want to make the enhancement of the E4 to move from trivial to very hard for the common user, I haver no doubt that they could achieve it. Nothing is un-hackable given time and access, but the effort to do so can increase exponentially to the point of it becoming only a mental game of chess between developer and hacker, and at great financial cost to both parties.

No one should be surprised, or annoyed, if FLIR do release a new 'hardened' build version of the E4. An ASIC based countermeasure would likely be enough to make life very difficult for a user to enhance their unit.

[Taucher]

After a ADDMENU hack the FLIR logo (in display down left corner) was removed, what should I do to add it again?
Thanks

See my footer how to edit/replace graphics inside the RCC file - just replace the logo image INSIDE the .RCC archive.

[Taucher]

Is it possible to make the fine adjust distance menu display in feet instead of meters?

The value is directly linked to the internal registry - didn't test if the value is in Ft when the whole cam is set to the imperial system.
In that case ignore (or edit yourself) the "meters" if you see them - that's hardcoded in the hack.

Just check with the regular focus distance adjustment option (what it shows), the added one is simply a more precise version.

[mrflibble]

No one should be surprised, or annoyed, if FLIR do release a new 'hardened' build version of the E4. An ASIC based countermeasure would likely be enough to make life very difficult for a user to enhance their unit.

An ASIC based countermeasure would also likely be enough to make it very difficult for Flir to keep the E4 economically viable. Cat, meet mouse. Mouse, meet cat. Have fun!

[steve1515]

Is it possible to make the fine adjust distance menu display in feet instead of meters?

The value is directly linked to the internal registry - didn't test if the value is in Ft when the whole cam is set to the imperial system.
In that case ignore (or edit yourself) the "meters" if you see them - that's hardcoded in the hack.

Just check with the regular focus distance adjustment option (what it shows), the added one is simply a more precise version.

Your new menu stays in meters regardless of how the meter is setup. So your menu label is correct. I was just hoping it would switch to feet also.

[cuda12]

Also think of the flood of returns that some will do when they think they have bricked their E4 (even if not covered) that alone would make flir enact some type of change to the firmware.

[mikeselectricstuff]

There is nothing to suggest that they are or are not taking action on this matter already.

..but not with any urgency. There is no question that they could easily have closed off the current "entry points" with minimal risk, and minimal need for testing, but the fact that many weeks have gone by and hundreds of units shipped since they must have been aware shows that it is not a high priority.

No one should be surprised, or annoyed, if FLIR do release a new 'hardened' build version of the E4. An ASIC based countermeasure would likely be enough to make life very difficult for a user to enhance their unit.

There would be no need for any hardware change. There are multiple things they could do on the existing
hardware to lock it down very effectively, if they really wanted to.

Without any inside knowledge, the only things we know for certain are that they are not so concerned about it that they are in any hurry to change the current situation (from which we can probably reasonably infer that a framerate hack is probably not viable) , and that they have sold a number of additional units that they wouldn't otherwise have sold as a direct result of  hackability.

If anything, I suspect Fluke are more pissed off than Flir.

[plesa]

For info only....

Junktronix has kindly sent me some high quality Thorlabs Germanium lenses to experiment with.

I shall see if they have an application with the E4. These lenses will be superior in performance to ZnSe if their characteristics permit their use. They appear to be suitable as a primary lens with a back focus distance of around 15mm

http://www.ebay.co.uk/itm/111231059741?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1423.l2649

Datasheet

http://www.thorlabs.de/newgrouppage9.cfm?objectgroup_id=1780&pn=LA9410-F

I will advise on their usefulness to us as soon as I have done some tests (hopefully this week)

I'm gong to buy it if your test will be sucessfull, thanks for taking action in this matter :-)
Am I right that germanium should give more consistent FOV measurement than ZnSe?

[Taucher]

@Plesa:
germanium has a higher refractive index than ZnSe - but FOV (field of view) does not depend on that factor - more on the lens geometry.

Are you talking about image distortions or the depth of focus?

[tomas123]

Guys, I am working on my batch version of Tomas123 flir.php.

I plan some kind of histogram - which temperatures in which count are populated over image.

Ideas ?

Hi Daves,

after reading http://www.imagemagick.org/Usage/files/#mpr I found a better solution

first create an grayscale image and select any palette (replace the auto-level with the ImageMagick fx operator)

Code: [Select]

exiftool -b -RawThermalImage withlens.jpg | convert - gray:- | convert -depth 16 -endian msb -size 320x240 gray:- -auto-level image.png

exiftool -b -Palette withlens.jpg | convert -size 224X1 -depth 8 RGB:- -separate -swap 1,2 -set colorspace YCbCr -combine -colorspace sRGB -level 4096,60928 palette.png
check -colorspace RGB or sRGB for better colors (Linux/windows)

then write a histogram with this one-liner -> no temporary files and no pipe 

Code: [Select]

convert palette.png -resize 256x200! image.png -define histogram:unique-colors=false -write histogram:mpr:hgram mpr:hgram -delete 1 -alpha off -compose CopyOpacity -composite hist_color_trans.pngor with background color white

Code: [Select]

convert palette.png -resize 256x200! image.png -define histogram:unique-colors=false -write histogram:mpr:hgram mpr:hgram -delete 1 -alpha off -compose CopyOpacity -composite -background white  -alpha remove -alpha off hist_color_trans.png

[Bonocr]

After a ADDMENU hack the FLIR logo (in display down left corner) was removed, what should I do to add it again?
Thanks

See my footer how to edit/replace graphics inside the RCC file - just replace the logo image INSIDE the .RCC archive.

  // FLIR logo
        Image {
            id: logo
            source: "../images/Sc_Logo_FlirHardEdges.png"
            visible: greenbox.appState !== GreenBox.FacetArchiveView

Is this the point that I have to modify?
In what way?
Thanks

[Taucher]

After a ADDMENU hack the FLIR logo (in display down left corner) was removed, what should I do to add it again?
Thanks

See my footer how to edit/replace graphics inside the RCC file - just replace the logo image INSIDE the .RCC archive.

  // FLIR logo
        Image {
            id: logo
            source: "../images/Sc_Logo_FlirHardEdges.png"
            visible: greenbox.appState !== GreenBox.FacetArchiveView

Is this the point that I have to modify?
In what way?
Thanks

Sc_Logo_FlirHardEdges.png <- that file got edited (into totally transparend) in the hack's RCC archive - you'd have to re-package the rcc file once you exchanged the file back to the original one - you could even replace it with a logo of your own
I've updated my post about RCC extraction (see links in footer) to include extracted versions of the archives as re-packing seems to be less a problem compared to unpacking.

[plesa]

@Plesa:
germanium has a higher refractive index than ZnSe - but FOV (field of view) does not depend on that factor - more on the lens geometry.

Are you talking about image distortions or the depth of focus?

These lenses has the refractive index reduced from ~4 to ~2 by AR coating. The transmission is lower 90% compared to 97% for ZnSe.
Of course the FOV depends only on the geometry.
Yes, I mean the differencies in the measurement should be lower when you are using Ge, than ZnSe. It can be due to the chromatic dispersion.
This can be eliminated by some calibration I think.