Zebra Technologies leverages AI to bring deep learning to its Aurora software

“Manufacturers across many industries face longstanding quality issues and new challenges with advances in materials and sectors such as automotive and electronics.”

by

Zebra Technologies has revealed that it is utilising artificial intelligence (AI) features to enhance the product offering of its machine vision software platform, Aurora. With this latest update, the digital solutions provider aims to provide its customers with a software solution for complex visual inspection processes.

“Manufacturers across many industries face longstanding quality issues and new challenges with advances in materials and sectors such as automotive and electronics,” said Donato Montanari, Vice President and General Manager, Machine Vision at Zebra Technologies. “They are looking for new solutions that complement and expand their current toolbox with AI capabilities needed for more effective visual inspection, particularly in complex use cases.”

Aurora’s new capabilities will aid companies within the automotive, electronics, semiconductor, packaging, and food industries in particular. With no-code, deep learning optical character recognition (OCR), drag-and-drop environments, and extensive libraries, the platform will be able to solve complex problems that traditional, rules-based systems would have previously struggled with.

Improvements made to the Aurora platform:

  • Aurora Design Assistant: With new deep learning object detection and an upgraded Aurora Imaging Co-pilot companion, users will be able to utilise a new, dedicated workspace for training a deep learning model on object detection.
  • Aurora Vision Studio: The new update provides users with a new training engine with upgraded training data balancing capabilities that will provide better results for low-quality datasets. Additionally, training is now quicker and repeatable, with the deep learning add-on now compatible with Linux systems for inference only.
  • Aurora Imaging Library: Now featuring expanded capabilities, the library has anomaly detection tools to conduct defect detection and assembly verification tasks to find abnormalities. This training is unsupervised with only normal references required.