There's a lot of hype around artificial intelligence, and it can be hard to sort the signal from the noise.

So why not talk to a real live human? Give us your info and we'll contact you right away.

Contact Us


Announcing Our New DNNCam

We’re proud to announce our first major product release, the DNNCam™.

The new DNNCam has been in development for more than a year, the culmination of a process that started with a simple goal:

Develop a deep learning camera system for remote and/or hazardous environments that can operate with minimal power and bandwidth requirements.

We think the DNNCam accomplishes this goal perfectly.

What Makes The DNNCam™ Special?

Front view of DNNCamMost camera systems for use with deep neural networks are “dumb.” Which is to say, most camera systems do not have processing capability on-board. This means that these cameras often require substantial infrastructure to deploy.

A camera used in a retail foot traffic counting system, for example, might require:

  • High-speed internet access, either via WiFi or (very often) via a standard wired network connection
  • A standard 110VAC power connection
  • A mounting location that’s free of moisture, dust, and heat

While these requirements might be easily met in a laboratory, they often come with a hefty investment in the real world.

Deploying a traffic counting system in a retail store with multiple entrances, for example, could require hundreds of feet of network cable, new electrical wiring, a protected mounting point with some sort of cabinet, and a crew to install all of these things over several days. The total investment could be in the tens of thousands of dollars.

Outdoor mall

Counting and classifying shoppers at an outdoor mall is a technical challenge for most camera systems, to say nothing of the expense of providing electrical connections and network access.

But if the store deploys the DNNCam instead…

  1. DNNCam can be installed with a battery system and a small solar panel, eliminating the need to run electrical wiring to each camera mounting point.
  2. Because DNNCam has on-board processing, it can digest what it sees and spit out the essential data using a low-cost cellular phone network connection.
  3. Because DNNCam is dustproof and waterproof, it can be mounted without concern for the elements.

DNNCam™ Is Built For The Real World

DNNCam has an IP67 rating, which means it can be get wet – or left outside in a dust storm – without being damaged. The camera also uses very little electrical power, which means it can be operated almost indefinitely on a battery pack and a solar panel.

Fish ladder

Want to count the number of fish jumping up a ladder, sorted by species? DNNCam can do it, running on solar power and transmitting minimal data over an analog cellular network.

Because DNNCam has potent on-board processing capabilities, it can be programmed to process what it sees, distill the resulting data, and send back only the data you’re trying to collect. Unlike most AI camera systems, the DNNCam does not require a high speed internet connection, cloud-based data storage, etc.

Instead of producing terabytes of raw data that requires processing, DNNCam produces kilobytes of data that you can plug directly into your applications and reporting.

DNNCam™ Is Ready For Your Application

DNNCam is compatible with most major existing AI software applications. If your company already has a trained neural network, you can upload that system to DNNCam and start gathering usable data.

Additionally, Boulder AI offers systems development and consulting services. If you want our help integrating DNNCam into your existing systems, we can do that.

To learn more about the DNNCam’s specifications and pricing, click here.

Darren Odom

The founder of Boulder AI, Darren’s engineering resume includes consulting for leading-edge research companies, product design and development, custom software application development, and more. Darren holds numerous patents, and is a recognized expert in the fields of deep learning and AI hardware development.