Yet another cheap thermal imager incoming.. Seek Thermal

[ricksastro]

When I ordered my Android one a while back it took them forever to ship it and my charge hold disappeared, then they proceeded to charge the amount again, but it wasn't a double charge since the original charge never went through.

Look in your bank history statement to make sure you are not charged twice, but I did find it confusing they way they did the charges.

I was charged $199 on 10/21 (charge for the actual unit) and then again on 12/11 (the screw-up)

[eneuro]

Hopefully, there is easy solution for not wanted double charges for the same thing-use prepaid master card and transfer money when need to pay for something.
Internet is a jungle-you might be tested if you notice double charge for nothing 
Had many attempts to charge my prepaid master card-the most spectacular one was on new petrol station-they tried charge not calculated equivalent in USD but requested... local currency amount, so wanted 3x more money, but failed because transfered only expected amount of money to this card 

[iRad]

Just received notice from Seek that the iOS camera has finally gained approval from Apple. I have now been able to reorder with the offered 10% discount and free shipping.   

Let's see how long they take to deliver now...

[heavybarrel]

The app has been approved for some time now, this go around it has been hardware approval for the use of the lightning connector, the key component they had a shortage of.  Some might be interested in the goings on of my day today.  I got the same email that everyone else got, "We can now ship ios devices" and "use the code to purchase at the 10% off." I still did not get
my refund from my original purchase so i fired off a cordial email. I actually had a response in about a half hour! Go figure!
A person by the name of Ryan replied that "the last batch of refunds/reversals were submitted over the weekend and are expected to settle with card issuers later today." He also said that if i don't see a credit by tomorrow i should let them know so they can investigate further.  I can say that the fast response is a good sign in light of everything else. I don't expect overseas shipping any time soon, they're still scrambling to fulfill domestic orders. Pretty sure they lost control so much that they had to cancel all ios orders simply to figure out where they were in that mess, aka start fresh. Hopefully they can get it right this go around and everyone in the forum can finally get the gadget in hand! 

[iRad]

The app has been approved for some time now...

As I first reported in this thread almost two months ago.
« Reply #264 on: October 18, 2014, 11:53:01 PM »

Sure did take a long time for the camera itself (lightning connector) to be approved.

I also received the credit for my initial purchase just today. Looks like Seek may finally be on a roll.

[heavybarrel]

Fingers crossed irad!! Lets hope for the best!! With everything being sorted with apple there should be no more problems! Best of luck on the new order, hoping mine goes well to!

[Fraser]

No revised software release for Android yet though. It will be interesting to see how well the iOS application works with the SEEK. Surely they have addressed the issues that have been highlighted here by now ?

Could this turn out like the Hantek offerings....nice hardware, pity about the software 

Decent software is every bit as important as the clever hardware that provides the data.

Aurora

[iRad]

We should know soon enough...    or 

[mikeselectricstuff]

No revised software release for Android yet though. It will be interesting to see how well the iOS application works with the SEEK. Surely they have addressed the issues that have been highlighted here by now ?

Could this turn out like the Hantek offerings....nice hardware, pity about the software 

Decent software is every bit as important as the clever hardware that provides the data.

Aurora

A complication is that AIUI any update requires re-approval in the Apple store, but it's not like they've not had time to get stuff sorted.

[JcDenton]

Bought the cheapest znse lens I could find on ebay, $15 shipped.  Seems to work ok.

[frenky]

I've bought cheap 50mm and 100m ZnSe lens too but haven't played with it yet.
How did you mount it on the seek module and at what distance?

Meanwhile I've found probably the easiest way for making vertical & horizontal pano head.
I was looking at kossel/delta 3d printer...
I'll use this small steppers:


3 steppers will be instead of vertical bars of kossel and the platform for seek module will be where printer extruder is.
This way I'll be able to easily move seek vertically, horizontaly and in circular (axial precession) movement.

[JcDenton]

I've bought cheap 50mm and 100m ZnSe lens too but haven't played with it yet.
How did you mount it on the seek module and at what distance?

Meanwhile I've found probably the easiest way for making vertical & horizontal pano head.
I was looking at kossel/delta 3d printer...
I'll use this small steppers:


3 steppers will be instead of vertical bars of kossel and the platform for seek module will be where printer extruder is.
This way I'll be able to easily move seek vertically, horizontaly and in circular (axial precession) movement.

I haven't found a way to mount it, was going to make something from acrylic tube or similar. It has a 25mm focal length right up against the seek no gap.  50-100mm would probably be better to give a bit wider shot, I was looking at a 7.5" focal length but the the seller ran out.  I'd kill for a 3d printer, but just can't afford one, I love 3d modelling so I do want to get one in the future if they come down in price.  A cheap desktop cnc would be great too.  The ultimate would be a metal printer that could print metal right into the plastic print.   Something like this but all materials at the same time -

[efahrenholz]

Seems like this thread has gone cold. Nobody with any firmware decoding experience has stepped in, and the windows development has basically gone on hiatus. Also, it appears the android app that promised gradient reversal but broke temp readings has also disappeared.

I wish I could offer some help but I've got little experience with current programming languages. I am proficient with arduino code, but that doesn't translate at all to this stuff. I do have some electronics experience and I have most the major heavy tools to do any investigation but no experience on where to look for anything. This is just beyond my skill level to hack into. I guess I'll start learning how to tap into the communications lines. I figured Mike would have already raped this project to death by now.

[miguelvp]

It's just a bit busy at work and home, holidays and all that. Maybe after Xmas I get some free time.

[cynfab]

And there is more than Window$, my python code will run on OS's other than Linux. I still haven't gotten as good a banding reduction as miguelvp or frenky, but it does work, and in real time.

   ...ken...

[Fraser]

With regret I have sort of parked the SEEK on a shelf awaiting better software. So no lens investigations or other hardware investigations yet.  Since I bought the SEEK I have purchased a Ganz C-Allview dual technology (thermal/Visible light) PTZ and a FLIR SC500 scientific thermal camera. These have all my attention at the moment. The software to drive them works and they produce decent images. The SEEK remains in the background though, so I have not given up on it yet.

Aurora

[frenky]

I still haven't gotten as good a banding reduction as miguelvp or frenky, but it does work, and in real time.

This works ok for me:

Code: [Select]

//tempFrame= calibration frame - image frame + 10000

//then:
int iPeak = 10230;//depends on the image, I don't have it hardcoded
int iRange = 570;// -||-
int minPix = iPeak - (iRange/2);
int imaxPix = iPeak + (iRange / 2);
int maxPix = imaxPix;


//put 207th colum into array:
for (int y = 0; y < 156; y++)
{
                arrLinet[y] = tmpImgLine[206, y];
}

//get min & max values of the array:
minTune = arrLinet.Min();
maxTune = arrLinet.Max();

//range is:
range = maxTune - minTune;

//"floor" the array so the min is 0 and max = range
for (int y = 0; y < 156; y++)
{
arrLinet[y] = arrLinet[y] - minTune;
}
maxTune = maxTune - minTune;



//tune lines:
double tuningStrength=0.003;
           
            for (int y = 0; y < 156; y++)
            {
                for (int x = 0; x < 208; x++)
                {
                    pixVal = tmpImg[x, y];//current pixel

                    if (pixVal < minPix) pixVal = minPix;
                    if (pixVal > imaxPix) pixVal = imaxPix;
                    maxPix = imaxPix - minPix;
                    pixVal = pixVal - minPix;
                   
                    pixPrescaler = 1;
                    pixPrescaler = ((double)pixVal / (double)maxPix);

                    linePrescaler = ((double)arrLinet[y] / (double)maxTune);
                    newVal = tmpImg[x, y] * linePrescaler * tuningStrength * pixPrescaler;//higher the value, less correction is applied
                    tmpVal = tmpImg[x, y] - Convert.ToInt32(newVal);
                    tmpImg[x, y] = tmpVal;
                }
            }


[mikeselectricstuff]

Seems like this thread has gone cold.

My main interest was investigating connecting the sensor direct to the sensor, which looks rather tricky, both electrically and the amount of processing needed, and along with the disappointing image quality means I haven't been doing much with it.
Might do an extreme teardown at some point.

[eneuro]

Nobody with any firmware decoding experience has stepped in, and the windows development has basically gone on hiatus.

I haven't had  a chance now destoy Seek unit and make its simplified circuit diagram with its MPU pins marked and scope waveforms.
This is not such a big deal to disassemble this code, like most of people there probably do to ensure quality of code generated by higher level compilators like C and optimization influence on output code.
There is no presure to hack this hardware, while this thing can be usable without this hack, but I'm pretty sure there are many people around globe who probably downloaded already his own "improved" code, but can't make this public avalilable
Commodore 64 was programmed by many EEVBlog members using.... hex editor and without any compilator, just by putting program code bytes into its RAM, so changing Seek firmware is piece of cake,  by end of the year has much higher  priority
I believe in I'm still ahead with gradient reduction in Seek dongle, while Seek failed and showed nothing.
Anyway circular motion has higher priority  than firmware hacking, while it allows make significant improvement in output image quality in some apps, so simply no need for the moment to mess with hardware details-this dongle is accessible with a few lines of C++ code and for the moment it is fine to automate many processes which require thermal objects detection 

[Uho]

Making circular motions we reduce the amount of non-performing pixels. But the gradient is no reduced. I spent tests. Alluminum tube with a lens creates gradient. If you put the flap behind the lens - no gradient. Maybe you have ideas how to make not expensive shutter behind the lens. Area curtains should be larger. Apply the original does not work.

[Fraser]

Someone is selling a close-up lens for the SEEK camera on ebay.

Price isn't bad considering parts cost and the need for a 3D printed mount. I have no connection with the seller.

http://www.ebay.com/itm/NEW-SEEK-THERMAL-Macro-Lens-CAMERA-IMAGING-ANDROID-FLIR-LWIR-NIGHT-VISION-/201245366108?pt=US_Other_Cell_Phone_Accessories&hash=item2edb28975c


Correction: they are using sealed cell high density foam and not a 3D printed plastic holder. A clever idea as I have used such foam before and it is tough stuff and fitted to the task. Expensive to buy thick sheets of it though. No idea how they created the required profile for the lens and SEEK parts though. Looks a neat job.

Aurora

[JcDenton]

Someone is selling a close-up lens for the SEEK camera on ebay.

Price isn't bad considering parts cost and the need for a 3D printed mount. I have no connection with the seller.

http://www.ebay.com/itm/NEW-SEEK-THERMAL-Macro-Lens-CAMERA-IMAGING-ANDROID-FLIR-LWIR-NIGHT-VISION-/201245366108?pt=US_Other_Cell_Phone_Accessories&hash=item2edb28975c


Correction: they are using sealed cell high density foam and not a 3D printed plastic holder. A clever idea as I have used such foam before and it is tough stuff and fitted to the task. Expensive to buy thick sheets of it though. No idea how they created the required profile for the lens and SEEK parts though. Looks a neat job.

Aurora

Interesting, I think you would just need a hole saw, and 2 drill bits.  I have some different foams lying around so I might try this.
Edit, just made one from 1" acrylic tube with 1/16" thickness.  It is slightly bigger than the seeks diameter, so I used some velcro on the inside so it stays tight against the seek and also holds the lens in place.

[mole]

Found this interesting doc on the seek thermal on non android platform..
http://centmesh.csc.ncsu.edu/ff_drone_f14_finals/Video/finalReport.docx

Oh, hi btw, my first post here :p

[iRad]

Just got my iPhone Camera in. Does not look like any of issues have been fixed yet.

Picture of crowded Apple Store.

Picture of escalator in the Shopping Mall. Persons easily identifiable on the second floor about 150 feet away from where I stood when taking the photo.

[Spyke]

If they had just placed the shutter on the outside of the lens this would be a non-issue. Maybe FLIR has a patent for that too...