Together with selected hardware partners, Robovision AI offers a proven end-to-end solution for Automatic Defect Classification (ADC) in the semiconductor wafer-production process. Here’s how the Robovision platform unlocks your inspection machine’s full potential.

Deep learning for Defect Classification

Robovision AI tackles the challenge of traditional computer vision systems in quality assurance by turbocharging it with Deep Learning. Our specialised semicon algorithm enables automated wafer-map pattern classification, eliminating the need to train the algorithm for each layer. That means that the more diversified your production is, the bigger your gain will be. It also translates to a faster time to market and the fast switching of production lines.

Higher yield at higher speed

We enable manufacturers to timely detect and classify flaws in the production process. Thus, the number of Defective Parts Per Million (DPPM) can be significantly reduced and they can increase production yield as a result. Our platform does so at high speed with near-perfect accuracy score: Robovision AI has proven to be able to achieve +95% accuracy in ADC at 40,000 multi-view defects per hour.

A unique solution for end-to-end AI creation

Training your AI model is the first step. Maintaining it and monitoring the performance is what separate Proof Of Concepts from Industry Grade Solutions. Our Robovision AI platform manages the entire AI lifecycle and makes every step as smooth and intuitive as possible, from data management to model deployment. Moreover, since the platform doesn’t require coding skills or AI knowledge, it is actually the operators who will make your machines smarter than you thought possible.