The In-Sight 2800 vision system from Cognex (Natick, MA) combines deep learning technology with traditional rule-based vision tools to solve a wide range of inspection applications. It is designed for tasks from presence/absence detection to complex categorization and sorting problems.
The In-Sight 2800 Series can catch small, subtle defects with accurate, deep learning-based error detection capable of solving OK/NG applications and classifying parts with variation based on multiple defect types or user-defined features. It features a 1.6 MP and field-interchangeable optical accessories to increase flexibility and solve variable applications. It has a compact form with straight or right-angle configurations and a multi-core processor as well as indicator lights for operator feedback.
The In-Sight 2800 is programmed using the EasyBuilder interface within the In-Sight Vision Suite and is also embedded with a full suite of traditional and deep learning-based vision tools that solve a variety of error-proofing applications. The library accomodates all skill levels by providing access to rule-based algorithms such as Measure Distance, Pixel Count, Count Patterns, and Math and Logic Tools. It additionally features Cognex’ ViDi EL tools that leverage deep learning-based technology to learn in real time. ViDi EL identifies and sorts parts based on multiple features or characteristics, and it is trained in minutes, using as few as five to ten images per class with no coding required.
For more information: https://www.cognex.com/products/deep-learning/in-sight-2800-vision-system