Optimizing process by using AI for detecting product errors and poor quality at high speed. Overview of the outcome of a process line and sorting low-quality products from high-quality products.
Camera for detection is already installed. Requires a lightweight, but very fast edge computing solution for running the recognition algorithm and updating the sorting mechanism.
Real-time image analysis
- Run image recognition algorithm in less than 100 msec
- Take action (send action) to sorting mechanism in real time
Collect data and update AI
- Collect data for historical usage
- Train AI-algorithm
Integrate 3rd party systems for management and control
Crosser Node deployed in a Virtual Machine:
- Connecting to machine camera processing box.
- Running image recognition locally o Sending anomalies to Machine Execution System (MES)
- Sending notifications/errors to Machine Engineers and Enterprise System
- Sending data to AI training in the cloud
- Crosser Cloud:
- Customer organization set up with roles
- Resources library setup