Innodisk's EMPA-I101 and EGPA-I201-C1 are high performance machine vision units (VPUs). They are powered by Intel's third generation Movidius Myriad X Vision Processing Unit (VPU) processor, which is designed to run neural networks.
The VPU application accelerates video stream processing and helps create a compact solution for face recognition, vehicle license plate identification, production control on conveyor lines and much more.
With the right combination of EMPA-I101- and EGPA-I201- modules with a central processing unit, you can get a 30x improvement in AI performance.
Improving the capabilities of machine vision
Embedded industrial PCs are poorly suited for image recognition and processing. Their CPU is usually selected based on power consumption and heat dissipation, making them unsuitable for running AI applications.
&Like how a graphics card takes the load off the CPU to render graphics, the VPU helps speed up image and video processing in AI applications. Innodisk Artificial Intelligence Accelerator Modules allow you to do this with minimal heat generation and keep your PC's power consumption low.
Integrating AI into IoT Applications
Innodisk AI accelerator cards are available in two form factors: mPCIe with one Myriad X core (EMPA-I101) and M.2 2280 with two Myriad X (EGPA-I201). They support Windows and Linux operating systems. The Intel OpenVINO toolkit allows you to work with deep learning frameworks: Caffe, TensorFlow, Apache MXNet, and Open Neural Network Exchange (ONNX).
The EMPA-I101 and EGPA-I201 AI accelerators from Innodsik enable you to use Artificial Intelligence in Internet of Things (AIoT) applications. Moreover, the system will not only be reliable, but also affordable.
