Project 136: Thermal Imager Project Update 8: FLIR Lepton + Raspberry Pi 2
It may not be possible to use an Arduino device with the Lepton. Even the ARM based Arduinos, such as the Due, may not be powerful enough to properly read data from the Thermal Imaging Camera.
The reason for this is that the Lepton will go “out of sync” if the data is not read from it as fast as it is being sent. The lepton also sends data at 27 frames per second. Legal restrictions require the lepton to have a maximum refresh rate of 9 frames per second, so the lepton actually duplicates frames, meaning only one frame in three is unique. However, despite this, all 27 frames need to be read per second, or the lepton will essentially stop transmitting data.
I didn’t know this, and I planned to simply read the 80×60 pixels at the maximum rate that I could, and then store them in memory. This does not seem to be possible.
The Raspberry Pi, however, given it’s significantly higher specs, could easily handle this. In addition, due to the fact that it is essentially a fully-fledged computer, (An “SOC”, or System on a Chip, to be precise) I could do may other things with it too. I could add other devices, such as a camera sensitive to the IR spectrum, and a laser illuminator. That means I could have both near IR and far IR imaging on the same device. I could also add a conventional visible light camera, so that I could combine this with the Lepton’s imaging in order to increase resolution, the same way that the Flir One does it.
The Raspberry Pi is cheap, and is actually about the same size as the Due, maybe even slightly smaller, so it would definitely fit my project’s size. It also runs off of 5V, the same as the Arduino. The only issue is that it seems like a shame to waste such a powerful device on one project.
The Raspberry Pi fulfills essentially the same role as the Arduino Tre, which I have been waiting for for some time now. I had been planning to see if I could get a Dedicated Server for T3D running on the Tre when it came out, I could use the Raspberry Pi 2 for this instead. This will be a future project.