Universally Accessible
Universally accessible to allow any engineer, without any AI knowledge, to be at the center of the setup process, with simple and intuitive steps straight at the production floor, onsite.
Self-supervised
Our product is software only, designed to support any camera via a standard GigE connection and to support all leading PLCs for fast automation support.
Retrofit
It supports your existing investment for quick and affordable retrofits or to your next new inspection line. Designed to run on premise, on any standard industrial computer.
AI engines
Our goal is to offer a fast and scalable solution for quality assurance. We have four AI engines working together to create a runtime model. Through the power of these engines we are able to process untagged data straight from the production floor. Our approach allows the user to guide the model creation process, reflecting the way the inspected product is used and inspecting what matters and where it matters.
QualiCenter
QualiCenter™ is our central guidance cockpit, where the quality manager interacts with the solution. You can define multiple regions of interests (ROI) as needed, define any defect type, define sizes and thresholds for any defect, in any ROI, to reflect the way the inspected product is later used and to the specification it needs to conform to.
We have designed QualiCenter™ to be the center of knowledge transfer between the years of experience of the quality manager and the model creation process. After all, nobody wants a black box. This is the reason both humans and machines are crucial – this is Augmented AI.
Instead of sorting huge amounts of images, labelling and creating separated data sets for training and testing, QualiSense automatically does these tasks for you. We are therefore able to reduce the time required to deploy a fully functional inspection station by orders of magnitude.
A high number of images is required to obtain a good representation of the product, but the work involved isn’t necessary. It is not about reducing the number of images, but rather automating the process involving a high number of images so the amount is transparent to the user.
This website uses cookies to ensure you get the best experience on our website Got it!