FLIR E4 Wifi Resolution and Menu Hack Thread

[blaqdragun]

Thanks Dave. Yes I can confirm the 3.12 update worked perfectly. Super interested in the advanced features for 3.12. Looking forward to Bud's future releases. Thank you Bud and everyone who contributed.

[Bud]

Thanks DaveWB and  blaqdragun. You can go back to page 20 and download the Advanced pack for 3.12.0.

[blaqdragun]

Darn. I sincerely apologize.  I ran this hack and commented without completely inspecting my unit.

I believe I spoke too soon regarding the basic at least for my install...I assume the shooting menu should include more than the following:

Recording Mode: Only has Single Shot
Thermal Menu (2 Stacked Windows Icon): Is disabled
Measurement: Only has No Measurements and Center Spot
Color Menu: Seems to only have 1 row of the basic palettes namely: Iron, Rainbow, Rainbow High Contrast, White Hot, Black Hot, Arctic, and Lava
Temperature Scale: Only has Auto.

I confirmed the resolution hack using Exifdata.com
MakeFLIR Systems ABModelFLIR E4 WifiExposure Time1/40 (0.025 sec)Focal Length6.5 mmFile Size16 kBFile TypeJPEGMIME Typeimage/jpegImage Width320Image Height240Encoding ProcessBaseline DCT, Huffman codingBits Per Sample8Color Components3X Resolution72Y Resolution72Software3.12.0YCbCr Sub SamplingYCbCr4:2:0 (2 2)YCbCr PositioningCenteredDate and Time (Original)2018:01:26 09:57:39Color SpacesRGBDigital Zoom Ratio1CompressionJPEG (old-style)OrientationHorizontal (normal)

I retraced my steps, but still got the same results. Any ideas? I assume I should not move forward with the Advanced if the menu items for the basic are not appearing, correct?

Edit#1
DaveWB does your menu appear correctly?

Edit#2
Reverified the image resolution using http://exif.regex.info/exif.cgi.  The extracted image says 80x60.  Rechecked my setup and I overlooked the fact that I am running a Win7 64bit laptop, not a 32bit one.   

I assume this resolution failure is due to using the 64bit machine to run the python script, correct?  Does anyone have information on running a VM for this liberation?  Again I apologize for the misinformation.

[mrubbert]

Known rabbithole is that one of the old conf. or common is still there in camera.
One way is to delete first then copy if you are to lazy to use stopapp.
check file date to be shure copy is done in camera.

There is a good easy phython explanation in this thread with picture.

[mrubbert]

This is a very nice ad Bud: "toolbarSeparators" value="true"
Better overview. Thanks.

Can 2.0 owners dream about getting higher temerature messurement ?

[caipiao]

Edit#2
Reverified the image resolution using http://exif.regex.info/exif.cgi.  The extracted image says 80x60.  Rechecked my setup and I overlooked the fact that I am running a Win7 64bit laptop, not a 32bit one.   

I assume this resolution failure is due to using the 64bit machine to run the python script, correct?  Does anyone have information on running a VM for this liberation?  Again I apologize for the misinformation.

using Python 32bit on 64bit Windows seems to have worked for me....

[Bud]

Known rabbithole is that one of the old conf. or common is still there in camera.
One way is to delete first then copy if you are to lazy to use stopapp.
check file date to be shure copy is done in camera.

There is a good easy phython explanation in this thread with picture.

yes make sure to delete the original dll and conf, then copy the new ones and double check the timedate stamp on them.

[Bud]

@blackdragun: you have classic symptoms of the config not loading. redo your dll and config as said above.

[Bud]

Can 2.0 owners dream about getting higher temerature messurement ?

not as i know of.

[Fraser]

The Ex series do not contain the required additional pixel bias voltage tables or associated calibration files for the other temperature ranges.

The temperature imaging capabilities of the Ex series may be extended using various attenuators placed in front of the lens. A non multi coated UV (Haze) glass photographic filter enables imaging of a gas flame etc. Sadly calibration is not simple though. Other materials can act as attenuators as well. It is a case of experimenting. Try plastics first.

Fraser

[David C]

Hi, new member here. First of, a huge thanks to Bud, Fraser, Mikes, and many others who greatly contributed to the pertaining TIC threads here and leading to the E4 "upgrade for personal learning" liberation, as well as deeper understanding of how TIC and thermography works trough experiments and vulgarisation of the matter.

Just like another member here, I wish I could donate to Bud, but unfortunately doesn't use Bitcoin or any block chain platform. I do have paypal and every other money wiring means tho, so I'll be happy to send some your way Bud if you ever provide another mean of gifting you.

I bought my E4 wifi 2L 3.9.0 in early december 2017, calibration card said sept. 27th 2017. Yesterday I got around to do the full upgrade with the basic and advanced packages as well as the preset_threeSpots mod from Squawk. Then I spent the next 24 hours discovering even more of what's around me, including some deceiving ghost hunting in a huge cemetery at midnight in -18° C (I guess ghosts don't come out when it's that cold ?) and go some gorgeous pictures of trees against the clear sky.

I've read most if not all of the related threads to the E4 on here in the last months and felt confident about the procedure. It did help that I'm computer inclined and know my way around, however I really had trouble like many others regarding the step about the Python command line. I think it would help to provide a bit more detailed insight on how to perform the step for those who aren't familiar with this. I'll explain how I made it work for be in my next post below.

Thanks again to everyone for your time and efforts, it was thrilling to first explore with the base E4, now it's truly amazing to witness the world with a higher resolution and more versatile palettes and measurements options.

Also worth pointing that you can record video and remote control the Flir camera using the free PC-based Flir IR Camera Player (http://www.flir.com/instruments/display/?id=50428) from Flir, as I haven't seen any mention of this yet in the "can I record video" discussions here.

[David C]

From Bud's Readme file in the 3.9.0 Basic Package :

...
4. Obtain cfccfg.zip archive:
   https://www.eevblog.com/forum/thermal-imaging/flir-e4-thermal-imaging-camera-teardown/msg596959/#msg596959
   and extract cfccfg_V2.py from it.

5. Install Python 2.7 . The script may not work with later Python versions.

6. Open the supplied conf.cfg , scroll down to the very bottom to the line that starts with "# ID " and replace the 9-digit number with your camera serial number.
   Save the file.

7. Encrypt the updated conf.cfg from step 6 with your suid by running this command from command line on your PC with Python:

   python cfccfg_v2.py XXXXXXXXXXXXXXXX conf.cfg conf.cfc
                    (where XXXXXXXXXXXXXXXX is the suid string from step 3)

   This step will produce a new encrypted conf.cfc .Save it to your working folder for transferring to the camera.
...

So after you've copied-pasted the URL of step 4 in your web browser, you'll reach this post :

Just a small update on cfccfg.py for 2.3. It now drops the signature/tail from the .cfg file. You can also make changes to  the .cfg in plain text before converting back to .cfc, the file tail is created with the proper file size info (the signature is set to all 0). Note that you need the "HIRES" patch  first in order to skip the CRC/signature check for this to work.

I tested adding a few lines to the .cfg file (comments, putting back 80x60 resolution...), convert to .cfc and use that on camera. Seems to work fine.

original post:
https://www.eevblog.com/forum/testgear/flir-e4-thermal-imaging-camera-teardown/msg594600/#msg594600

The attached cfccfg_V2.zip file is the one you need. Download it and then unzip the folder. Inside the folder, you'll find a cfccfg_V2.py file and you'll have to move or copy it to the Python main folder that was created upon the installation of Python (on my computer the Python installer default path was directly to the main C: drive, but you can specify another path like the Program Files folder). I decided to keep things simple and have the Python folder directly on my C: drive.

After reading on this thread that people where having the same issues with syntax and line1, and Squawk saying something about the .cfg file path that had to be specified, I started trying combinations of everything with the PC command line, the command prompt, the Python command prompt, etc. So here's what to do :

I moved the conf.cfg file of step 6 to my C: drive, then ran the following line from the PC command line that's available in Start Menu/Run.

python cfccfg_V2.py XXXXXXXXXXXXXXXX C:/conf.cfg conf.cfc

I did replace the XXX with my SUID obviously.

Once you click "Ok", nothing seem to happen, but if you look into the Python main folder, you'll see that a new file (conf.cfc) was created ! Simply copy that file back to your working folder where you keep your other files to be transfered (I copied mine into the basic 3.9.0 folder where the conf.cfg file was taken from).

The best way to get it right is simply to copy-paste the command line of step 7 into a new text file along with your SUID and then replace the XXX with your SUID and add the proper path to the conf.cfg file (easiest way is move the .cfg file to your C: drive and add C:/ to the command line).

Hopefully that helps. I've posted screenshots below.

[mrubbert]

The Ex series do not contain the required additional pixel bias voltage tables or associated calibration files for the other temperature ranges.

The temperature imaging capabilities of the Ex series may be extended using various attenuators placed in front of the lens. A non multi coated UV (Haze) glass photographic filter enables imaging of a gas flame etc. Sadly calibration is not simple though. Other materials can act as attenuators as well. It is a case of experimenting. Try plastics first.

Fraser

That is great  help,
i do not need any exact numbers on higher temperatures. Most offen the interesting is to see the difference.
Plastics i will try
I am still learning about this after 100 and 100 and 100... pages.....
Only in one building is so many variabels. One flore heating was crystal clear, other was smear out in cement ofcource and so on.   ( lettit cold down and fire up the heating is one way ).

High heat i was thinking about is automotive catalytic converters. Maybe not that importent, can still messure the o2 values and the backpressure that tells the condition anyway or better way.

I like this tool 

[Bud]

You may pull something together using a attenuator filter as Fraser suggested, and using the External Window Compensation feature in the Advanced pack. That would be interesting to see if with that combination higher temperature can be measured with acceptable accuracy.

[Bud]

I think it would help to provide a bit more detailed insight on how to perform the step for those who aren't familiar with this.

the easiest way is probably just put the cfg in the Python folder and run the command from there. This will eliminate the need to specify paths.

[Bud]

Also worth pointing that you can record video and remote control the Flir camera using the free PC-based Flir IR Camera Player (http://www.flir.com/instruments/display/?id=50428) from Flir, as I haven't seen any mention of this yet in the "can I record video" discussions here.

For some reason that player never worked for me thru Ethernet. Also, as far as I remember it does not display radiometric video (i.e. with temperature measuring cursor so you can measure different parts of the image), it is just a color overlay. For true radiometric video streaming we made an effort here but it was a quite convoluted solution and only available on Linux.

[blaqdragun]

I restarted from scratch and stopped at the dll step.  Can someone please let me know if the attached image is the full 320x240?

[Bud]

No it is not

Just the DLL step alone does not enable resolution. You must do your config.

[blaqdragun]

I am not sure the cfc file is created correctly.  The attached file has lost the thermal resolution information and my shooting menu is crippled again. 

I followed the steps correctly. Perhaps my SUID is wrong?  In step 3 you ask to get the SUID by two different means.  Why is that?  We expect them to be the same right? I copied the SUID to the cfg file and encrypted it with the "python cfccfg_v2.py XXXXXXXXXXXXXXXX conf.cfg conf.cfc"
script. 

I then delete the old cfc and upload my newly created one.  I apologize if i am missing something obvious. 

Edit#1 - I am running Python 2.7.14 (x86 version) on Win7 64bit.

[David C]

Also worth pointing that you can record video and remote control the Flir camera using the free PC-based Flir IR Camera Player (http://www.flir.com/instruments/display/?id=50428) from Flir, as I haven't seen any mention of this yet in the "can I record video" discussions here.

For some reason that player never worked for me thru Ethernet. Also, as far as I remember it does not display radiometric video (i.e. with temperature measuring cursor so you can measure different parts of the image), it is just a color overlay. For true radiometric video streaming we made an effort here but it was a quite convoluted solution and only available on Linux.

The IR Camera Player software allowed me to control the camera via USB and mirror the camera display to the computer as well as saving still images and recording videos. All adjustments and settings were done using the camera settings, except that you could use the keyboard instead of the camera buttons. That would allow for an easy remote control of the camera (let's say you're logging data from inside a car with the camera mounted outside of said car to avoid sticking your hand and head out of the car since the windshield or windows are blocking the view) and a cheap and easy way to make thermal recordings.

[Bud]

I followed the steps correctly. Perhaps my SUID is wrong?  In step 3 you ask to get the SUID by two different means.  Why is that?  We expect them to be the same right? I copied the SUID to the cfg file and encrypted it with the "python cfccfg_v2.py XXXXXXXXXXXXXXXX conf.cfg conf.cfc"
script. 

Does not seem you followed the steps correctly. You do not copy the SUID to the cfg file. You copy the camera serial number to the cfg file - please re-read the instructions. Do not confuse S/N with SUID, they are totally different.
Two different means to read SUID is just that - 2 different means. they give the same SUID.

[blaqdragun]

Changing my username to moron at the time of this writing. 
Of course, everything worked perfectly as soon as I actually followed the directions Bud meticulously assembled. (Reading is fundamental...) I am the proud new owner of a liberated E4 wifi!

I need to look into a way to compensate Bud for all the time spent on this project and with me personally.  Thank you very much to you (Bud) and everyone who contributed to make this liberation a reality.

Ok no more wasting your time.  Happy thermography-ing!

[caipiao]

trying to realign MSX after liberation but can't seem to get the CRC32 computed...

I am using the script from here: https://www.eevblog.com/forum/thermal-imaging/flir-e4-thermal-imaging-camera-teardown/4375/?action=post;quote=406423;last_msg=1405463

and computed on the original calib.rsc (minus the last line) and got a different result vs the one in the last line. Anyone know if they are using a different algo now? seems unlikely but a bit stuck....

[Bud]

What is wrong with adjusting MSX using the menu?

[caipiao]

thanks Bud for replying; nothing wrong - I didn't know you could...

Just tried though: won't allow me to adjust left and right (or I am missing something else  )... I attach a misalignment example.