Cloud-Powered AI Lab Enhances Accuracy and Efficiency in Vision Metrology
With its latest machine vision release, wenglor sensoric group has introduced AI Lab: a browser-based platform for training AI models for industrial image processing. The AI Lab enables efficient data management and AI model training in the cloud, while the uniVision 3 software enables image recording and the high-performance execution of AI models on machine vision products. The weHub software ensures seamless communication between the AI Lab cloud platform and the uniVision 3 image processing software. In the ‘AI Loop’ workflow, weHub handles the upload of image data and the download of AI models. The AI Loop provides users with an intuitive workflow for easy and fast AI modeling.
With uniVision 3, the modular standard software for wenglor machine vision devices such as the B60 series smart cameras and the MVC series machine vision controllers, images and height profiles can be evaluated easily and flexibly – without any programming knowledge. With the uniVision AI license package, rule-based applications can now be combined with self-created AI models to ensure maximum flexibility in image processing applications. wenglor machine vision hardware is automatically detected in the local network via weHub and all relevant device settings can be made directly. In the AI Loop, weHub also serves as an interface between the AI Lab cloud platform and the local network.
Maximum Ease of Use and Exceptional Performance
Working in the AI Lab is characterized by a uniform interaction pattern and logical workflows. The weHub connection tool enables images to be uploaded quickly and easily to the cloud automatically from uniVision 3 or via file upload. Several data sets can be efficiently managed to create AI models directly. State-of-the-art cloud infrastructure with computing power in the high-performance range enables deep learning for very powerful AI models for image classification, which makes it possible to train even large data sets and complex AI models.
Scalable User Management and Targeted AI Model Optimization
The cloud structure enables flexible, location-independent access to the AI Lab, so projects can continue at any time. User management enables scalable working: Records can be edited simultaneously and shared between different users. The data stored in the cloud is backed up regularly to keep it available in the event of technical failures. Various tools are available for evaluating and optimizing created AI models. In addition to the integrated heatmap function, both automatic validation and a review function can be used for targeted improvement of the model performance.
For more information: www.wenglor.com
