RoadTest: Intel® Joule™ 570x Developer Kit
Author: armour999
Creation date:
Evaluation Type: Independent Products
Did you receive all parts the manufacturer stated would be included in the package?: True
What other parts do you consider comparable to this product?: null
What were the biggest problems encountered?: null
Detailed Review:
Unboxing
I am going to break this RoadTest for the Intel Joule 570X into a few blogs and update the upper portion of the review. I am so happy to extend my knowledge of Computer Vision using the Intel Joule. I have almost completed a camera trap with the Raspberry Pi and Raspberry Pi camera. The most impressive achievement was to identify the breed of the animal captured with motion detection using OpenCV , Watson Node Red and the Visual Recognition node. I plan to do the same with the Intel Joule but probably in a very different way.
My first impression was how small the board was. The Intel Joule has built in WiFi and Bluetooth .
The back of the box had a nice summary of specs for the compute module.
It also included a Heat Sink and a USB Type 3 Cable . There was also a 16 SD card and some screws and a pamphlet that included a reference to a web site:
https://software.intel.com/en-us/intel-joule-getting-started
I needed to pick up a few items referenced in the website. I had a difficult time finding the power supply in town so I just ordered one with the required specs from the Robot Shop. I also researched the cameras and decided to order the Intel RealSense Camera (R200) Developer Kit. Both the power supply and camera should be here early next week. Intel has included some robust documentation so I know this will not be an issue. One article I find interesting is:
https://software.intel.com/en-us/articles/using-librealsense-and-opencv-to-stream-rgb-and-depth-dataUsing LibRealSense and OpenCV to stream RGB and Depth Data
My next Blog should be completing the setup of the workstation and preparing the camera.
Setting Up the Workstation
The power supply arrived but the RealSense camera will not arrive until June 14th.
The first step was to connect the Intel Joule to my laptop and make note of the COM port. I already had the FTDI drivers and Putty installed from a different project. I also had a USB Cable Type B . First power up the Intel Joule:
Green lights are always good 
I'm using an Acer Laptop with Windows 10 installed.I opened up Device Manager and looked for Ports (COM & LPT). Under USB Serial Port I could see the device was assigned to COM10.
I opened up Putty and configured the settings as shown:
Once I performed my first boot I was taken to a command line. To check my version I entered:
cat /etc/*-release
Now to connect to WiFi. The platform uses Connman. So at the command line I entered:
connmanctl
connmanctl> scan wifi
connmanctl> services
connmanctl> agent on
connmanctl> connect wifi__managed_psk
Below is a screenshot of my connection:
I picked up a micro HDMI cable to connect the platform to my RCA TV. That also works quite well. When I get home from work I want to try :
startxfce4
This should bring up a desktop environment with a file manager and other tools. My next steps is to install Ubuntu and the RealSense SDK.
Installing Heat Sink
Before the camera arrived I noticed the board getting warm, So I installed the Heat Sink that came with the kit.
https://www.intel.com/content/www/us/en/support/boards-and-kits/000022314.html
Installing Ubuntu
The starting point for the next steps is:
https://developer.ubuntu.com/core/get-started/intel-joule
Update the BIOS
The steps can be found here.
https://software.intel.com/en-us/flashing-the-bios-on-joule\
The pictures and screen shots were helpful in completing the upgrade. The process ended in my screen showing the correct result.
Creat a Ubuntu SSO account and generate a SSH key
After I created the account , I then looked at the instructions to generate the SSH key. As I am using a Windows 10 laptop, I downloaded Git for Windows to run the commands. Below is the result after I generated the key:
The next step was to open and save the id_rsa.pub file in Notepad. The contents was entered in the following page and saved.
Hardware needed for Install
The documentation recommends a USB hub which you will need to install Ubuntu. My Manhattan USB Hub worked well. The only piece of hardware that did not work was the micro HDMI cable I had brought for this project. Last weekend I started to hear popping noise on the monitor. Sometimes the monitor would shut down . After the store replaced it , I have had no issues. I also have a dongle for the mouse and keyboard , a couple of USB Flash drives and the R200 USB camera !
The Camera did arrive and it looks awesome. It comes with a USB Type C cable.
I have the Ubuntu Core16 image for Intel Joule downloaded and the image has been written to a USB Flash Drive. The Ubuntu Desktop 16.04.1 LTS has been written to a second USB flash drive.
I plan to finish the Ubuntu Desktop to night and to-morrow. Then I will be setting up the RealSense repository and installing samples. And if I am lucky the camera will be ready to go tomorrow !
Using the Realsense Camera and a WebCam
Yes I know that the Intel Joule has been discontinue. But I am still interested in Computer Vision but I am still going to continue. Since I last updated I had to fight abit with the install of Ubuntu. This turns out to be more with an issue of my WiFi which I begged the owner of the apartment to fix and was able to download Ubuntu to the Intel Joule. The next challenge was trying to use the RealSense R200 camera in a USB Hub. This did not work in a USB 2 or USB 3 hub. I found a forum from Intel suggesting that the camera should be in the USB 2 port. So that meant I had to switch to Bluetooth for the keyboard and mouse. My friends at work donated a keyboard and mouse that worked liked a charm. The board also seemed to continue to have an issue with overheating when the camera was plugged in. Although I had the heat sink installed that came with the kit you could feel how hot the board became with the camera attached. So....I added a little fan from work and that seemed to control the heat.
I installed OpenCV from Adrian's amazing tutorials that can be found here. I have used his tutorials to install OpenCV on the Raspberry Pi and find his instructions to be clear and informative. I did not think it would work with the R200 but it did work with a cheap USB webcam I had in my box of electronics. This is a screen shot of running the python code:
The Frame Delta = difference between the original first frame and the current frame The Thresh frame shows motion (with a black blob). The Security Feed shows a live feed of the camera.
Using the R200
So the WebCam worked with OpenCV but not with the R200. I continued to run some code with the RealSense Until I had the WiFi under control, the examples would cut off the WiFi. Below are some examples :
