Yet another cheap thermal imager incoming.. Seek Thermal

[Trax]

Could someone explain me how "destroying" pixels in software helps you bypass some patents, to me that sounds like bovine excrements?

[Fraser]

@Trax,

You would have to ask the FLIR lawyers. FLIR took SEEK Thermal to court over patent infringement. This pixel deletion requirement was the result. You should be aware that the two men behind SEEK worked for Indigo Thermal  who were bought by FLIR.. There is a lot of complex Patent stuff happening in the background as a result.

Aurora

[frenky]

I was playing with macro lens for my Seek and funny thing happened:



I even prefer it that way was since it will be easier to make lens holder...

[Trax]

@Trax,

You would have to ask the FLIR lawyers. FLIR took SEEK Thermal to court over patent infringement. This pixel deletion requirement was the result. You should be aware that the two men behind SEEK worked for Indigo Thermal  who were bought by FLIR.. There is a lot of complex Patent stuff happening in the background as a result.

Aurora

Stange, do you have any more details like for example how many pixels are deleted or in what pattern? Its just to satisfy my curiosity

Cheers
Trax

[frenky]

I've posted it a few posts back:

That white pattern are disabled pixels (ignore the one top left that doesn't fit the pattern)

[Fraser]

@Trax,

Every 15th pixel is inactive. SEEK Thermal claim that the combination of 12um pixel size and dead pixel correction overcomes these pixel loses.

Aurora

[miguelvp]

Yup, every 15 pixel is blanked on purpose to avoid patent infringement according to Seek themselves in this post buried within this thread:
https://www.eevblog.com/forum/testgear/yet-another-cheap-thermal-imager-incoming/msg540041/#msg540041

[z01z]

There's a new version of the Android app, it says to have improved image processing.

[Fraser]

From memory, SEEK Thermal always said that their app would be subject to ongoing improvement as the initial version was virtually a BETA release to get the product out the door.

Many of the issues that were experienced with the camera when it was first released could be dealt with in the app if the appropriate techniques are applied. As we have seen from the excellent work of Forum Members here, the camera is capable of producing a decent image. I only hope that SEEK Thermal have the skill set in the company to extract the very best performance out of the hardware via the phone apps.

They need to have a strong response to the release of the new FLIR One in order to maintain market share. They still have the advantage of greater temperature range.

Aurora

[frenky]

Previous version:


Current version:


and the winner is: jadew 

[bsliv]

I got an email from Seek with a request for participation in Seekware SDK pre-release.  Maybe innovative 3rd party apps will help Seek against Goliath.

The sign up page has the Windows SDK scheduled for release on 07/15/2015.  Better late than never.  I signed up. 

[miguelvp]

I got an email from Seek with a request for participation in Seekware SDK pre-release.  Maybe innovative 3rd party apps will help Seek against Goliath.

The sign up page has the Windows SDK scheduled for release on 07/15/2015.  Better late than never.  I signed up.

Thanks for the heads up, that made me check my personal e-mail and sure enough there it was

[frenky]

I got the mail to, but this Pre-release is for US people only. 

[lod]

Hi,

I've been reading through the fantastic work done by people on this thread but understanding it took a long time. As the thread approaches 2000 posts there is a huge amount of information not to mention the referenced material.

What I have done, for the next person that tries this, is translated my notes into a public wiki.
https://github.com/lod/seek-thermal-documentation/wiki

The wiki is currently a blend of information and references to useful posts in the thread.

I didn't want to steal the information in here or deny people credit for their hard work, so blocks of knowledge or data are referenced. Facts that were scattered through posts I consolidated and summarised directly in the wiki.

The wiki is public, anyone with a free github account can edit. I encourage the contributors here to look through it and update any information that they think is missing or incorrect.

My hope isn't to supplant this thread, rather provide a long term consolidated reference for information posted here and elsewhere.

I am also attempting to reverse engineer and document the USB protocol of the device. As I view this documentation as a more formal document it resides in the github repository itself. I also welcome contributions to this but want to use pull requests retain a bit more control over the process.

[frenky]

Wow. Nice job. 

[barehill]

Thanks Aurora, yes quite interesting! Seek has begun selecting candidates for their SDK, at last.

[barehill]

Thanks Aurora, yes quite interesting! Seek has begun selecting candidates for their SDK, at last.

[Ben321]

I'd be interested in seeing some unprocessed pics from the Seek thermal imager. Preferably uploaded either as an attachment here or to the image hosting site Imgur, or to any place that does not re-encode them, but rather hosts the image files exactly as uploaded. Thermal imager pictures often have 2 pictures in the same file, one is a standard JPEG (for giving you a visual impression of what was captured by the imager) and embedded in that file is the raw (or losslessly compressed) actual thermal data recorded by the pixels of the thermal imaging focal-plane-array (usually with a bit depth of more than 8 bits per pixel, which requires 2 bytes per pixel to store the data). That is why I would be interested in seeing a 100% unaltered copy of some of the images it takes. In the case of the Seek thermal imager, these unprocessed images will have a resolution of 206x156 and will have a file size of no less than 63kb (the file size needed to store the 2byte-per-pixel uncompressed image data with a resolution of 206x156, assuming that it is not losslessly compressed raw data), so I can tell if it's processed or not, by looking at the image resolution and file size.

From there I should be able to write my own program that will extract the raw thermal data. This will give me the true ability to compare its image quality to other similar thermal imager camera's (like the Flir One, with a 160x120 thermal focal-plane-array). Note, that to compare quality I must use the raw thermal image, not the JPEG image, as the JPEG image combines the thermal image with a higher resolution visible light image (in the case of the Flir One, the device has it's own camera, and with the Seek Thermal imager, it depends on the phone's camera). I'm looking forward to get a thermal imager for my cellphone, and I need to compare the actual raw thermal images from the Seek Thermal imager and the Flir One imager, in order to determine which to buy.

So anybody who goes to this forum would be helping me out a great deal, by posting an unprocessed original image taken by the Seek Thermal imager, and also the same for the new Flir One (assuming you have access to both, otherwise I'll have to wait for more than one person, each with one of these imagers, to post their images here). I have noticed that there are absolutely no original images from the Seek Thermal camera, nor from the new 160x120 Flir One, available on the net anywhere, and I've already searched Google Images, and all I can find are promotional images on the websites for these devices. These promotional images all have been processed, as they all lack the embedded raw thermal pixel data.


And one more request. In case I do get the Seek Thermal imager, I would like the desktop PC application made by "sgstair" called Seeker, that I see a lot of use from in this thread. Can somebody please compile it for me, and post the EXE file here as an attachment (or post to a file hosting site like Mediafire and link to the download here)? The guy named "sgstair" only offers the source code, not a compiled EXE file at his Github site. I don't have the version of Visual Studio needed to compile it, so I hope somebody else who has compiled it can make available the EXE file for me to download.

[frenky]

You can't get unprocessed images from Seek android app. All you get is 824x624 jpeg.
The best thermal image you can get is from this app:
https://www.eevblog.com/forum/chat/seekthermal-how-to-correct-the-image-received-from-the-sensor/msg696895/?topicseen#msg696895

[encryptededdy]

I'd be interested in seeing some unprocessed pics from the Seek thermal imager. Preferably uploaded either as an attachment here or to the image hosting site Imgur, or to any place that does not re-encode them, but rather hosts the image files exactly as uploaded. Thermal imager pictures often have 2 pictures in the same file, one is a standard JPEG (for giving you a visual impression of what was captured by the imager) and embedded in that file is the raw (or losslessly compressed) actual thermal data recorded by the pixels of the thermal imaging focal-plane-array (usually with a bit depth of more than 8 bits per pixel, which requires 2 bytes per pixel to store the data). That is why I would be interested in seeing a 100% unaltered copy of some of the images it takes. In the case of the Seek thermal imager, these unprocessed images will have a resolution of 206x156 and will have a file size of no less than 63kb (the file size needed to store the 2byte-per-pixel uncompressed image data with a resolution of 206x156, assuming that it is not losslessly compressed raw data), so I can tell if it's processed or not, by looking at the image resolution and file size.

From there I should be able to write my own program that will extract the raw thermal data. This will give me the true ability to compare its image quality to other similar thermal imager camera's (like the Flir One, with a 160x120 thermal focal-plane-array). Note, that to compare quality I must use the raw thermal image, not the JPEG image, as the JPEG image combines the thermal image with a higher resolution visible light image (in the case of the Flir One, the device has it's own camera, and with the Seek Thermal imager, it depends on the phone's camera). I'm looking forward to get a thermal imager for my cellphone, and I need to compare the actual raw thermal images from the Seek Thermal imager and the Flir One imager, in order to determine which to buy.

So anybody who goes to this forum would be helping me out a great deal, by posting an unprocessed original image taken by the Seek Thermal imager, and also the same for the new Flir One (assuming you have access to both, otherwise I'll have to wait for more than one person, each with one of these imagers, to post their images here). I have noticed that there are absolutely no original images from the Seek Thermal camera, nor from the new 160x120 Flir One, available on the net anywhere, and I've already searched Google Images, and all I can find are promotional images on the websites for these devices. These promotional images all have been processed, as they all lack the embedded raw thermal pixel data.


And one more request. In case I do get the Seek Thermal imager, I would like the desktop PC application made by "sgstair" called Seeker, that I see a lot of use from in this thread. Can somebody please compile it for me, and post the EXE file here as an attachment (or post to a file hosting site like Mediafire and link to the download here)? The guy named "sgstair" only offers the source code, not a compiled EXE file at his Github site. I don't have the version of Visual Studio needed to compile it, so I hope somebody else who has compiled it can make available the EXE file for me to download.

You can find raw FLIR One images here, straight from the app: https://www.eevblog.com/forum/testgear/new-flir-products/msg703078/#msg703078

They are full radiometric JPEGs and you can use the FLIR Tools software to turn off the image blending.

Alternatively, I've imported some raw FLIR One images into FLIR Tools (http://puu.sh/j03LM/e827998097.jpg) and taken off the MSX image blending (such that the images are thermal only). You can see them here: https://www.eevblog.com/forum/testgear/new-flir-products/msg704209/#msg704209

Since the FLIR images are fully radiometric JPEGs (meaning both the visual and thermal image is embedded inside), temperature data for each pixel is stored and the image blending, pallet, range etc. can be adjusted after the image is taken. I do not believe the Seek images are radiometric JPEGs - instead I think it just saves 2 images, one visual, one thermal. However I imagine with the Seek SDK you should be able to make an app that saves radiometric JPEGs or other metadata.

[Ben321]

I do not believe the Seek images are radiometric JPEGs - instead I think it just saves 2 images, one visual, one thermal. However I imagine with the Seek SDK you should be able to make an app that saves radiometric JPEGs or other metadata.

If they take a pair of images, can you please post up some of the thermal part of the pair? I'd be interested in looking at these.

And one more request. In case I do get the Seek Thermal imager, I would like the desktop PC application and driver made by "sgstair" called Seeker, that I see a lot of use from in this thread. Can somebody please compile it for me, and post the EXE file here as an attachment (or post to a file hosting site like Mediafire and link to the download here)? The guy named "sgstair" only offers the source code, not a compiled EXE file at his Github site. I don't have the version of Visual Studio needed to compile it, so I hope somebody else who has compiled it can make available the EXE file for me to download.

[marshallh]

Search for jadew's thread, he has a zip file with exe and such.

[frenky]

I posted the link to zip only a few posts back but he seems to be ignoring it:

The best thermal image you can get is from this app:
https://www.eevblog.com/forum/chat/seekthermal-how-to-correct-the-image-received-from-the-sensor/msg696895/?topicseen#msg696895

[Ben321]

I posted the link to zip only a few posts back but he seems to be ignoring it:

The best thermal image you can get is from this app:
https://www.eevblog.com/forum/chat/seekthermal-how-to-correct-the-image-received-from-the-sensor/msg696895/?topicseen#msg696895

Sorry. I didn't know that was a desktop application. You called it an "app" which usually refers to something on smart phones, whereas "application" refers to something on a desktop or laptop PC. Only after I downloaded and looked at it did I realize that it was actually a desktop "application", rather than an "app" as you had called it.

[miguelvp]

Four posts before yours:

https://www.eevblog.com/forum/testgear/yet-another-cheap-thermal-imager-incoming/msg707218/#msg707218

lod (first time poster and only post so far) has compiled this whole thread into a very usefull wiki:

https://github.com/lod/seek-thermal-documentation/wiki