Athena-A1 toolkits
ASRock Industrial’s athena A1 Kit is an AI edge camera that supports Intel’s Movidus™ and OpenVINO™ toolkits(Open Visual Inferencing and Neural Network Optimization) for sophisticated video and image analysis applications. It is more flexible operation without chassis design for Maker and software developer to implement recognition training models.
The athena A1 AI edge camera delivers 2MP resolution (1080p) with 30 fps,and a H.264 codec combining high definition footage and powerful video stream analysis capabilities.
Support for AWS IoT Greengrass allows for the deployment of cloud enabled inference functionalities that create value-added solutions for use-cases in theft/crime prevention or consumer behavior analysis based on machine learning.
Athena have installed two Edge computing software SDK：Intel OpenVINO Toolkit(Open Visual Inferencing and Neural Network Optimization) and AWS IoT Greengrass.Provide out-of-the-box AIoT development board,whithout tedious installation process. Plug-in has ready-made AIoT system. The system is x86 64bit CPU architecture and Ubuntu OS . Through the applications with Athena A1 Kit maker can develope application easily.
You can watch the Video to assemble the athena-A1.
Let me show athena A1 Kit spec：
The Intel NCS 2(VPU) enables deep neural network testing, tuning and prototyping, so developers can go from prototyping into production leveraging a range of Intel vision accelerator form factors in real-world applications.
Before use VPU to run demo. You must to run the batch file vpu_init.
The Athena-A1 system is default Ubuntu : 16.04 ,Openvino version :2019.1.144 . If you want to recover your system . You can download image to ghost to your athena-A1 .
Download the image file to your HDD or SSD. You have to use USB or other device to boot to Ubuntu. Open terminal and command as below:
Use root
sudo su
Ghost to the Athena A1
gzip -dc /imagefile_path/backupV008_customer.img.gz | dd of=/dev/mmcblk1
if it down you can see the information as below :
xxxxxx+0 record in
xxxxxx+0 record out
xxxxxxxxxx bytes (XXGB, XXGB) copied , xxxx s,21.0 MB/s
it mean the ghost was finish and need to restart the system
We provide some examples as below
You can follow demo to command .
This demo showcases Object Detection task applied for face recognition using sequence of neural networks. It detect age, gender, emation.
This demo showcases Pedestrian Tracking scenario: it reads frames from an input video sequence, detects pedestrians in the frames, and builds trajectories of movement of the pedestrians in a frame-by-frame manner.
Before Inference
After Inference as below
This demo demonstrates how to run the Image Segmentation demo application, which does inference using image segmentation networks.
Road-segmentation : This is a segmentation network to classify each pixel into four classes: BG, road, curb, mark.
Semantic-segmentation : This is a segmentation network to classify each pixel into 20 classes Road,sidewalk,building,wall,fence,pole,traffic light,traffic sign,vegetation,terrain,sky,person .... etc
This demo showcases Vehicle and License Plate Detection network followed by the Vehicle Attributes Recognition and License Plate Recognition networks applied on top of the detection results.
