A couple months back, I decided to buy an Avnet Ultra96 V2 board and make something interesting.  The Ultra96 is a compact, inexpensive platform with some amazing capabilities.

 

https://www.avnet.com/wps/portal/us/products/new-product-introductions/npi/aes-ultra96-v2?CMP=AMER-296-ULTRA96V2-XLX-MKTO-ENL-TALK1-201909&mkt_tok=eyJpIjoiWm1Nek5UZGpZMkV6TlRZeCIsInQiOiJBbXRwNXRkNVRaRXdWdDRxYlwvV00yT1M1N3k2TmJKVzBtb0xLXC9yR1Z6XC8wcDRONXdBeUJOMWI0OVRxamxWd2pvMTI0dFRxZFVnM2RjY29pWTNucEtSMmF5KzFGeFhTSVlrQjFhUzBQWHVDbElYazZwalpNa3plbG1VSTBKYTV1dCJ9

 

I decided to use a tool called Model Composer which allowed me to work with an example application, simulate it and verify performance all within an easy-to-use graphical environment.  I did this mostly to learn all the individual steps myself.

 

I keep reading about self driving cars so I picked a color detection example that processes streaming input data and detects street signs.   To show just how easy it is to work with Model Composer and target an Avnet Ultra96 board, I made a 4-part video series.

 

https://www.mathworks.com/videos/series/getting-started-with-the-avnet-ultra96-development-board.html

 

After going through some of the documentation, I was able to get my system working and picking out road signs with good repeatability.  Of course, the output from this project is a functional block that is running on Ultra96 hardware.  In a real-world system I would need to by a camera and a display too. 

 

Nonetheless, it was an interesting project and I look forward to doing more work with image processing.

John