I just picked up an Intel Coral USB Accelerator ($59.99) from Mouser (after waiting a few weeks) for exploring Machine Learning and object detection and recognition. I am very impressed with the performance, and look forward to working with it some more. While researching the Coral Edge device, I came across this video , and thought it perfect for the Udoo X86 platform due to its inclusion of Arduino support. Here is the description from the video: The Pixy2 is a low cost yet powerful camera that is capable of object recognition, line tracking, and simple barcode reading. The device is the latest iteration of the Pixy Cam, a project built by Charmed Labs in conjunction with the Robotics Institute at Carnegie Mellon University. With a variety of interfaces and lots of code libraries and samples, the Pixy2 can be used with an Arduino, Raspberry Pi, Beaglebone Black, or just about any computer, microcomputer, or microcontroller. In this video, I will show you how the Pixy2 works, how to hook it up, and how to train it using software called PixyMon which runs on Windows, Linux, and Mac OS X. I’ll then show you how to easily hook up your Pixy2 to an Arduino and run code to detect an object, lines, intersections, and simple barcodes. Thanks to its onboard processor the coding for Pixy2 is very simple. It provides a very easy method of adding vision to your next Arduino or Raspberry Pi project. Here is the table of contents for this video: 00:00 - Introduction 02:18 - Pixy2 Introduction 05:53 - Pixy2 Unboxing 08:02 - Using PixyMon 14:01 - Color Signature Training 20:32 - Arduino Hookup 21:48 - CCC Hello World Demo 27:44 - Line Tracking Intro 29:24 - Line Hello World Demo As always there is a detailed article on the DroneBot Workshop website, you will find it here - https://dbot.ws/pixy2. I have not researched the PixiCam other than finding their web site: https://pixycam.com/pixy2/. the PixyCam2 is available here at $59.95 https://www.sparkfun.com/products/14678 and "all the usual places." This may be old news to some, but I haven't seen it mentioned in this forum. My hope is that this thread may spark some interest in Machine Learning among the savvy members of this forum, and someone may getting it working with the builtin Arduino hardware support on the Udoo X86. There is little question AI is the future. Have fun. Larry
For some reason the link to the video was removed from my post: h t t p s : / / y o u t u . b e / 3 9 1 d X D j q z X A
Okay. Some additional info: PIXY2 links to Arduino via SPI. Technical specs Processor: NXP LPC4330, 204 MHz, dual core Image sensor: Aptina MT9M114, 1296×976 resolution with integrated image flow processor Lens field-of-view: 60 degrees horizontal, 40 degrees vertical Power consumption: 140 mA typical Power input: USB input (5V) or unregulated input (6V to 10V) RAM: 264K bytes Flash: 2M bytes Available data outputs: UART serial, SPI, I2C, USB, digital, analog Dimensions:1.5” x 1.65” x 0.6” Weight: 10 grams Integrated light source, approximately 20 lumens
