Lepton SPI to USB via interface?

[Scott Morris]

Is there a reasonably simple way to get the Lepton breakout board to talk to a PC USB port so that I can wrangle the data via Matlab?  Most of the material I'm finding presupposes using a RPi board to create a screen image, while I'm after the actual pixel values. 

Would a simple SPI to USB interface be a workable solution to this?

[sa...@pureengineering.com]

Personality I use the NUCLEO-F411RE to capture the image, and use an ftdi rs232-to USB running at 921600 baud to get the data to a PC.  I can get the full 9Hz framerate.  

I have a simple win32 program that can grab the frames over the serial port

https://github.com/PureEngineering/LeptonModule/tree/master/ThermalView

Also, the original original lepton development kit uses this http://www.totalphase.com/products/aardvark-i2cspi/ to get the spi data to a PC. Somewhat expensive for what you are getting, but does work. 

   

[Scott Morris]

Thank you!  

Here's an annoying follow-up question:   All pride of authorship aside, which of these two options do you think will get data into Matlab faster/better given that we're in a lab with modest (at best) programming chops?  

We're currently using a GridEye board from DigiKey and are getting good (8 x 8 thermopile-based) data there after dealing with some PIC/ftdi wonkiness, but if buying the pricey Aardvark interface and using the FLIR SDK gets the Lepton board going more quickly, it will be a better choice in the long run.

[sa...@pureengineering.com]

The faster option is always what you are most familiar with. I haven't used the Aardvark yet, so I can't say if that is faster or not. 

what PIC are you using (any pic32 should work), are you familiar with the 410-202P-KIT?  I was looking at it the other day, and it seems the breakout would plug straight into that, might be another option. 

[Scott Morris]

You may be right, but since we're not really familiar with any of them it's a toss up at this point - I've contacted FLIR to see what they think since they have the SDK, but the Nucleo might be the best so far.

[Scott Morris]

FLIR Advised going the RPi route which makes things more complicated rather than less, so I'm getting the Nucleo board and we'll see how it goes from there.  Thanks!

[tz]

You might try to modify my arduino code in another thread.  It runs the SPI at the full speed possible and it has a USB connection, but the USB is polled so might not be fast enough.  There isn't enough memory on the arduino, but there are some Atmel chips with 16k.

[Scott Morris]

Has someone gotten decent data output from plugging it into an ARM arduino, like this?    The advantage to us would be one fewer side trip into another IDE...kind of the original problem (modest programming skillz).

http://www.pureengineering.com/projects/lepton/IMG_20140912_154957.jpg?attredirects=0

[Scott Morris]

Okay, per your suggestion I'm into the Nucleo-F411RE (what a joy to work with BTW...I'm hooked) - as far as the FTDI RS232 to USB would this do?

http://www.adafruit.com/products/70

I don't see any "gotchas" here, but then that's where they get their name....

Thanks!

[sa...@pureengineering.com]

http://www.adafruit.com/products/70

that cable would work fine. 

one thing is that it won't directly plug into the nucleo. so I would also get some extension jumpers so you can remap it easily and cleanly. 

http://www.adafruit.com/product/826

[Scott Morris]

Thanks - lots of jumpers at hand already.  More dumb questions in due course I'm sure.