Ease machine vision AI integration with turn-key edge solutions

The all-in-one solution allows machine vision experts and developers to build, test and deploy machine learning (ML) models faster, while giving IoT solution teams and system integrators the tools to connect, stream and automate machine vision work.

“We’re all feeling the pressure of ramped up production during these unprecedented times. Machine vision AI is one of the quickest, easiest and most cost-effective ways to add automation into existing processes without disrupting current workflows,” said Kevin Hsu, Senior Product Manager of ADLINK’s IoT Solutions and Technology Business Unit.

“Our Smart Pallet solution, for instance, has proven to deliver a 90% reduction in traditional barcode scanning time while also enabling customised, built-to-order pallets with no changes to the production line other than the smart camera’s installation. Using the highly integrated NEON-1000 effectively decreases the footprint, reliability and compatibility issue and also saves maintenance efforts for customers.”

The Intel Movidius Myriad X VPU (vision processing unit) optimises ML models and can provide some of the highest AI computing power. Eliminating complex integration of sensor modules, cables and VPU modules, the NEON-1000-MDX Smart Camera simplifies the machine vision deployment process. With the pre-installed EVA (Edge Vision Analytics) SDK, users can deploy neural networks optimised by OpenVINO without extra coding or integration efforts using a wide range of ready to use plug-ins based on environment and system requirements.

The new all-in-one NEON-1000-MDX AI smart camera supports product classification and defect detection to maximise production efficiency in smart manufacturing, as seen with the award-winning ADLINK Edge Smart Pallet solution. Smart Pallet adds intelligence and automation to manual warehouse fulfilment operations such as receiving, bin picking, packing, shipping and worker safety.

Smart Pallet provides an end-to-end integrated system to connect new and existing equipment, capture multiple image data streams and apply the high performance processing power of the NEON-1000-MDX VPUs to enable machine learning and inferencing at the edge.

With the ADLINK Edge software platform developers can connect and integrate with any existing cloud, machine learning platform, neural network, industrial camera, machine vision system, piece of machinery and more regardless of vendor. ADLINK’s machine vision AI software can classify what it sees, become smarter over time and also create automation workflows.

“For instance, if a box contains the incorrect order on a conveyor system, it can send an alert to a conveyor’s sorter to divert the box to a re-check area,” said Steve Cammish, VP of ADLINK’s IoT Solutions and Technology. “We’re bringing software developers intuitive programming, automation and device management running on a powerful smart camera designed for AI workloads. The automation here is key- saving time, decreasing costs and reducing complexity.”