Neurala Vision Inspection Automation (VIA)
Software Brings New Technology to Visual Inspection

Our patented Lifelong-Deep Neural Network© (L-DNN) has significant advantages over machine vision and traditional deep learning technology.

Neurala VIA for Visual Quality Inspection

Neurala VIA software consists of two modules, Brain Builder and Inspector. It can be installed on any computer with a NVIDIA GPU or an Intel i3 or higher processor. Brain Builder enables creation of AI models, walking users (including those with little or no prior AI expertise required) through the steps to build a well-performing model. Once a model has been created in Brain Builder, Inspector integrates that model into the production line, connecting to any GigE camera and outputting the image analysis directly on industrial networks.

DOWNLOAD PRODUCT BRIEF

Advantages of Lifelong-DNN vs. Traditional DNN

VS.

Less Data Required

VS.

Instantaneous Learning

VS.

Directly on small compute edge

VS.

Continues to learn

Brain Builder™ Cloud-Based Platform

Brain Builder, part of the Neurala VIA product, is a cloud-based solution that enables rapid AI prototyping, allowing you to annotate and train at the same time. Because of L-DNN's accelerated learning speed, learning can occur as new image data becomes available, so precision gains are instantaneous. You can quickly iterate instead of having to wait hours, days or weeks to understand the results. This means you can easily test a proof of concept to gain a better understanding of the impact of adopting vision AI on any quality inspection process.

Neurala VIA Inspector

Neurala VIA Inspector allows for seamless integration on the factory floor, saving you time and money by allowing you to retrofit onto your existing production line. It collects images from virtually any GigE vision camera and provides outputs over an industrial network directly to the PLC controlling the applicable machine to run the vision AI models.

We Only Require "Good" Product Images

Neurala’s AI can be trained with only acceptable, or good product images, unlike traditional DNNs that need both good and bad examples. Working with good samples only, L-DNN produces an anomaly signal when an abnormality occurs. This results in tremendous time savings since defective products appear infrequently and randomly on the production line, and it would take a long time to collect their image data