We are true to the proverb that seeing once is better than hearing a hundred times. And that’s why in SOS electronic we also like to do it this way. If you are interested in embedded products and artificial intelligence, whether for work projects or just as technology, listen to and watch the attached videos. We put it together from the latest webinar focused on the Boxer-8150AI industrial computer.
Iris Huang, Product Sales manager from AAEON will show and compare individual AI@Edge Boxer products based on Intel + Intel Myriad X VPU processors, NvidiaJetson modules, HiSilicon HI3559A platform and Google Edge TPU.
It is not mentioned in the video, but the manufacturer AAEON also offers mPCle, M.2 2242 and M.2 2280. These are the modules based on the Kneron KL520 neural processor unit (NeuralProcessorUnit, NPU), which add additional possibilities of deploying neural networks on hardware with low energy consumption.
We focused on the features of industrial computers based on NvidiaJetson modules. We also briefly mentioned the features of the software package – Jetpack 4.3 and the distribution of AAEON ACLinux adapted for AAEON hardware.
Videos
SOS electronic webinar with AAEON Part 1 – Product Line
https://youtu.be/-iIqEpBGqds
Success stories
There are many useful tasks suitable for artificial intelligence. Deploying it on energy-efficient hardware has many benefits, but one is extremely important: a short response time.
We enclose ideas for the successful use of AAEON computers in everyday working life, but also during the COVID19 pandemic. The use of artificial intelligence makes it possible to determine whether a person has a protective mask and an adequate body temperature, and can thus be used effectively against the spread of a pandemic in crowded spaces.
Videos
SOS electronic webinar with AAEON: Part 2 – Success Stories
https://youtu.be/8WXkAo2pkk8
Tip
If you are interested in the topic, see and read our tips for similar articles:
Why implement machine vision and AI into a workflow?
Installation
AAEON product manager, Mr. OceanLu, recorded the whole process step by step on video and explained why he recommends a computer with Ubuntu 16.04. If you want to use a newer version of Ubuntu, due to the scripts used to load the operating system image, you will need to install Python 2.7. The whole process takes just a few minutes. A little interesting thing. Notice that Mr. OceanLu has an antistatic wrist strap to prevent damage to internal ESD components.
Videos
SOS electronic webinar with AAEON Part 3 – Flashing Image
https://youtu.be/Gb1d-aFBW1Y
Cheat sheet
Connect the power supply, monitor, mouse and keyboard to the Boxer-8150AI and turn it on. After booting the operating system, start the terminal and follow the enclosed “cheat sheet”.
First, compile the demo applications:
$ /usr/share/visionworks/sources/install-samples.sh ~
$ cd ~/VisionWorks-1.6-Samples
$ make
See the compilation result:
$ cd bin/aarch64/linux/release
$ ls
Run the application:
$ ./nvx_demo_feature_tracker
or
nvx_demo_motion_estimation
nvx_demo_hough_transform
nvx_demo_stereo_matching
nvx_demo_video_stabilizer, ect.
Videos
SOS electronic webinar with AAEON Part 4 – Demo Visual
https://youtu.be/vWuAUzQ
If you need more information or you are also interested in other embedded systems or products from AAEON, do not hesitate to contact SOS electronic at [email protected].
