Hot Camera Contest: Build A Battery Powered Thermal Camera

Here’s a challenge for all you hardware hackers out there. Peter Jansen has opened up the Hot Camera Contest on to use a thermal imaging camera in a battery-powered project.

The challenge here is simple. Use a Flir Lepton thermal imaging camera module in a battery-powered configuration. There’s a catch, though: this is a project to use the Lepton in radiometric mode, where the camera spits out an actual temperature value for each pixel. Yes, this is a documented feature in the Flir Lepton module, but so far very few people are using it, and no one has done it with a small, battery-powered device.

The rules for this challenge are to use the Flir Lepton 2.5 in radiometric mode using either the Raspberry Pi Zero W or ESP32. Any software in this challenge must spit out absolute temperature values in a text format, and there must be a demonstration of putting the Flir Lepton into low-power mode. There are two challenges here, one for the Raspi and one for the ESP32; and winner will be named for each.

Getting More from a Fascinating Sensor

The Flir Lepton is a tiny little thermal camera that’s been available to the Maker community for some time now, first through GroupGets and now through Sparkfun. For a pair of Benjamins, the specs are very impressive: the Lepton has a resolution of 60×80 pixels and everything is can be read over an SPI port. The Lepton gives any project thermal imaging, and the PureThermal board turns the Lepton into a USB device.

Peter Jansen is the creator of the Open Source Science Tricorder (yes, it’s a tricorder) which took Fourth Prize in the 2014 Hackaday Prize. You can understand how he became interested in portable, and we’re sure whatever project he has in mind for this battery-powered Flir will be awesome.

This really is a great example of what the community is capable of. The goal here is to create useful Open Source drivers for some very interesting hardware, and there’s some prizes to sweeten the pot. Peter has a $125 Sparkfun gift card on offer for each of the two winners. And the challenge of solving a tricky problem and making designs easier for others is a powerful motivator. Who doesn’t like a challenge?

36 thoughts on “Hot Camera Contest: Build A Battery Powered Thermal Camera

  1. So if I understand correctly, I first must invest (Sparkfun: $293,95 excluding shipment costs OR Groupgets: 397,00 excluding shipment costs) in order to “win” a prize of $125.

    Looks to me that the real winner isn’t the one who will be receiving the 1st prize…

      1. Sorry you feel that way. This is a pretty cool thermal camera. It would be great if people who need the function in their designs could use this without having to wade through all the same challenges as the last person.

        That’s the core of this challenge: can we pull together a great library for all to use? I think it’s unlikely that people will buy the part just to work on the library, but a lot of people already have one of these in their parts bins… Here’s a reason to pull it out and have some fun.

    1. open source hardware is driven by a community of contributors who enjoy discovery, a challenge, and giving back to the community by sharing their discoveries and work. it’s only by standing on each other’s shoulders that we have the incredible community resources we have now, and only by sharing our discoveries that we get to stand a little higher tomorrow. if you’re primarily motivated by financial incentives, then this likely isn’t a good fit — i am in no way anywhere near cash positive on the many contributions i’ve made, from the open source science tricorders, or open source computed tomography scanners, or the recent magnetic camera. the primary return you get as an open source contributor is seeing folks use your work, learn from it, and build amazing things with it — and that’s worth more than any financial prize. the small prizes here are just a token from me to say thanks, the real thanks will come from seeing a great many new projects enabled by these contributions using these two low-barrier-to-entry platforms (the ESP32 Arduino platform and Raspberry Pi Zero W).

      1. It’s a fair question though, you should be up front about what YOU plan to do with the winning design. If you’re planning on selling it for a profit, you should make that clear (not saying you are, but it would certainly be a good way to get people to do work cheaply).

        1. Small bounties are common with HAD and Make contests. It’s probably why it’s mostly simple engineering or reverse-engineering..

          I’ve seen plenty of contests that were way more shady and nobody said anything.. Kind of late to bring the game theory or street smart monotony…. Save it for wherever you buy your dockers from..

  2. Why in god’s great name ist this thing still so hilariously expensive?
    It’s out there for several years now and they still charge almost 300 bucks for it. Thats insane in my opinion.

    1. The same price will get a 320×240 microbolometer sensor ic from some companies, but it is only the sensor. You need the optical system and readout electronics, etc. So I think the price can be ok for 160×120 without hassle, it is a small market after all.

      1. There’s an argument to be made though that the market is small because prices are so high. Thermal cameras are *really* useful, but the few companies that make them are keeping prices higher than the wider market can bear.

      1. Depends on what you call “full flir”
        This thing is US$240 from Sparkfun, plus whatever for shipping.

        The genuine FLIR TG167 with the same 80×60 resolution is currently $377.59 from (US), with free shipping (for Trumpland residents).

        But if you’re willing to settle for 32×32, there are multiple vendors selling a complete imager with LCD for under $300, like Amazon ASIN B075GTHZV9 for $260.

        If you’re willing to surrender your phone to the display task, Seek Thermal is selling their device for about $175, also on Amazon: B00NYWAHHM for this one.

        If you still want to DIY more cheaply, the Grid-Eye 8×8 sensor is still there at Sparkfun for $40.

      2. Flir one (older version) cost me $200 about 2 yrs ago. Has 160*120 (4 times more pixels) and a VGA visible camera and connects to either an ios or android device (different versions). Sure, not open source but A LOT MORE HARDWARE for less $.

    1. Two remarks:
      1. What do you call affordable?
      2. You can buy modules for phones that work through micro-USB, so what you need is a raspi driver for one of those and you are done.

  3. The Flir camera for Android only costs $199 on sale, and is fully usable from the gitgo. I was lusting over one of these sensors, but now have all I need with the addon Flir One pro.

    Also for about the same price as this, you buy a low end camera from cheap suppliers of such.

  4. Flir’s primary market is b2b and govt. That has two implications.

    First, they’re not used to consumer markets or two-sided prosumer ecologies. So they’re moving cautiously and probably making mistakes that wouldn’t be mistakes in their primary line of business.

    Second, their gross profitability all stems from big institutional sales, and cheap high res radiometric interoperable sensors would endanger their price and requirements deliverables negotiations. They’re not going to hurt their main line of business for the sake of what is essentially a speculative long-tail side venture. Even when there are meaningful, significant difference between the high end and maker products, purchasing managers either won’t know or will feign ignorance and use it as an excuse to squeeze Flir.

    For startups and serious entrepreneurs, the ones who might productize the lepton into a volume oem product, the price is already very reasonable. An investment like this ensures that if you buy it, you really will do something with it. So no I don’t blame them. If they cut prices (at least initially taking a loss), they can’t unring that bell.

  5. Late to the argument, but why bother with building one unless you’re driven to reinventing the wheel? A FLIR C2 can be gotten new for $500, used for less and works as a 30 FPM web-cam, though they don’t like to advertise it.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.