Please ensure Javascript is enabled for purposes of website accessibility
Powered by Zoomin Software. For more details please contactZoomin

AVEVA™ Vision AI Assistant

Step 1: Create and train a skill

  • Last UpdatedFeb 12, 2024
  • 2 minute read

A skill is an Artificial Intelligence (AI) model created for a specific purpose. After launching AVEVA Vision AI Web Client, you can use one the following skill templates to create a skill:

  • Anomaly Detection – This skill is trained to learn what normal is, and then applies a statistical test to determine if the current image represents an anomaly. This skill is useful when there are plenty of images used to represent the normal state, and a vast array of negative states for which images are not readily available.

  • Discrete State Detection – This skill uses a Supervised Deep Learning algorithm to distinguish between two known states. This skill is useful when both states are known and images for both are available to train the algorithm.

To create and train a skill

  1. Launch AVEVA Vision AI Web Client.

  2. Select Add Skill.

  3. Select the skill type.

  4. Enter a name and description for the skill.

  5. Select Create Skill.

  6. From the side sheet, select one of the following options to import the image datasets:

  7. Train the AI based on the type of skill:

  8. Connect to one of the following monitoring sources; the live camera feed the skill uses to detect anomalies or classify into classes:

    • To connect to a live camera, see Network.

    • To connect to a webcam, see Webcam.

    • To connect to a dataset, see Dataset.

    • To connect to camera programmatically, see Programmatic.

  9. Select Next.

    The Review and Run AI page appears.

TitleResults for “How to create a CRG?”Also Available in