AIR sequence of AI inference techniques deal with numerous AI purposes from facial recognition to clever transportation and extra
Advantech has launched its complete AIR sequence of AI inference techniques, to handle numerous AI purposes together with retail facial recognition, AOI/imaginative and prescient inspection, robotic, clever transportation and extra. The choices embrace completely different featured designs to suit various AI purposes. Advantech additionally offers Edge AI Suite, a software program toolkit with easy-to-use graphical consumer interface (GUI) and pre-trained AI fashions to assist prospects allow AI inference on edge gadgets.
Constructed for Edge AI: Allows real-time intelligence on various purposes
AIR-series consists of AIR-100, AIR-101 and AIR-200 which integrates Intel Movidius Myriad X VPU, and AIR-300 which helps a PCIe x16 excessive energy graphic card. Powered by Intel Atom x7-E3950 CPU and one Intel Myriad X VPU, AIR-100 helps multi-4K show and is appropriate for interactive kiosks and different retail purposes with facial recognition features. AIR-101 helps DIN-Rail design and 12V~28V wide-range energy enter with Intel Atom E3940 Quad Core processors and two Intel Movidius Myriad X VPUs and is ideal for AGV and manufacturing unit automation purposes. For greater computing energy wants, AIR-200 is an Intel Core i5 platform with two Intel Movidius Myriad X VPUs and helps as much as 1080p video encoding, decoding and multi-channel processing appropriate for visitors monitoring, license plate recognition and car classification purposes.
The highly effective AIR-300 helps Intel Xeon E3/ seventh & sixth technology Core i sequence processor and a PCIe x16 high-performance graphic card to allow quick inference and on-premises coaching for robotics and optical inspection purposes.
Accelerating AI Deployment with Edge AI Suite
The AIR sequence comes with the Edge AI Suite software program toolkit that integrates Intel OpenVINO toolkit R3.1 to allow accelerated deep studying inference on edge gadgets and real-time monitoring of system standing on the GUI dashboard. The newest Advantech Edge AI Suite v1.2 consists of an optimized pre-trained Yolov3 object detection mannequin well-liked in surveillance purposes which delivers 2.8x higher efficiency on VPU than operating with OpenVINO solely. Prospects can use Edge AI Suite v1.2 to execute the Yolov3 mannequin with VPU+GPU heterogeneous acceleration to get 5x higher AI inference efficiency than utilizing a single system.