HemaVision: Smart thermal imaging with computer vision

I’m just curious what professionals’ feedback on this newly-announced product campaign is:

Fun for homeowners. Not close to a professional tool.

Thermography if used to find anomalies requires good equipment and the skill to “read” your images and this is not a camera a thermographer would show up on a jobsite with. Other than that looks pretty cool.

Yup, this should work just fine…

Well its in color anyway;-)

Yep both of them ;-):stuck_out_tongue:

Pretty much sums it up.

Shutterless 82x62 thermal diode imaging array This is part of the problem!! Well below minimum acceptable resolution requirements.

5 megapixel CMOS image sensor It has a 5 megapixel image sensor but only 5084 points of actual data from imaging array. Sounds impressive, but the actual results are still the same.

Good material to place in stocking of your teenager at Christmas… although they may not even think it is cool.

Ah you are too hard on it… I think it has three colors!! What in the world is one supposed to get out of that image. I would be embarrassed to own one.

Another camera for people who find the cost of a real IR camera prohibitive. This type of tool does more harm than good as it is being marketed as a diagnostic tool and its far from that.

Now were up to 3 under $500 IR cameras.
But wait**, if you order now we’ll give you another one for the same low price of $299.95:mrgreen:

I’m glad to see it’s not FLIR this time. Whew.

The HemaVision includes a shutter-less thermal diode array with unprecedented image sensitivity and the platform includes modular image processing and computer vision software components. Furthermore, it will also tell you when your pizza is done in the oven.


And then there’s this:
40x31 degree thermal FOV, 115mK NETD
9Hz frame rate

Hey guys, I’m the guy behind it, I’d be happy to talk about image quality or tasks or anything. I realize that many professionals really need that higher resolution. This is a more affordable option that beats anything under $1,000. In fact, there are professional imagers around $1,500 that have the same resolution as ours but because their sensitivity is so bad, the images are not as good as ours. What makes this one useful is the image sensitivity which allows for improved detectability. Higher resolution (as long as sensitivity is not compromised) will improve detectability, but it comes at a large up-front cost, which keeps a lot of people out of the thermal imaging market.

Hi Erik, congrats on your project, wish you great success and a long line of professional products down the road. How do you plan to compete with guys like Flir?

Please say Hi to Marius Popescu

Nothing against your product but it surely works against the professionals who offer real thermography reports.

If you need a thermographer someone with education and the best equipment is going to charge a fee and do qualitative and quantitative reporting, if you want less and don’t need an accurate report then your product is going to be popular with some inspectors and homeowners.

I compare this to some guitars, you can buy an accoustic guitar for $100. But if your making a living as a musician your gonna have something a lot better. The difference?
Playability- Tone- Features, a guitar is a guitar to someone who can’t play and is tone deaf but to people who demand better the difference is noticable.

Hi Marius! (and thank you David)

David, thank you! We plan to complement Flir by focusing on the software. Our goal is to open new markets with and for them, rather than competing with them. I think it would be nice to work with them one day, I like their attitude towards R&D.

Paul, thank you for your comments. I hope this doesn’t detract from professional thermographers, and I hope we can add value to what they do now. We want to offer features that are not available elsewhere and also be affordable enough such that thermographers could use the HemaVision as a secondary thermal imager that allows you to take advantage of the faster transfer of images and reporting. If you do thermography, what do you find to be the slowest part of the process in going from taking images through to delivering the report? Is there a point in the process that seems like it could be eased?


No parts of the IR process are slow or problematic as far as my using IR.

You will get the price shoppers but thats the market when you sell to amateurs. The better inspectors will hang in and continue with the expensive equipment because it does the complete job. Your camera will deliver images that may cause some users to interpret the data wrong but that can happen with a $10,000 camera also. Seriously though, do you think IR people will use your product as a backup?

I wish you well.

I agree with Paul. Not many slow parts that become an issue. I think the only improvement I would like to see would be WiFi transfer of images. Which I’m sure newer IR imagers may have, at least the 25k ones. Just like most of the higher end DSLR cameras now have. It’s more of a convenience thing for me. Not something that I have to have to improve my time on taking images or reporting