MAESTRO
Maestro will lead you from idea to deployment of a fully operational offline iDetect-4.0 agent.
iDetect-4.0 Development cycle
Thanks to this scheme we divide by 10 the lead time to put a fully functional model on the field compared to standard deep learning technologies.
SEE MORE DETAILS
At design phase, the designer uploads a sample dataset and setups the agent configuration
SEE MORE DETAILS
At evaluation phase, the designer tests the configuration on a test dataset. If the performances don’t reach expectations, it can go back to design and adjust the settings.
SEE MORE DETAILS
At deploy phase, DavinSy agents are built and packaged inside an iDetect-4.0 instance for each device in the target inventory.
SEE MORE DETAILS
At usage time, operator can monitor production health. iDetect-4.0 integrates with machine controller to receive status and pilot the machine by providing feedback. Operator can re-label erroneous predictions or newly spotted defects.
SEE MORE DETAILS
Thanks to the automatic model qualification, Quality service is informed of availability of qualitative models and can decide to deploy them or not. Deployment is done live without stopping the machine.