1. Zebra Aurora Focus User Guide
  2. Zebra Aurora Focus User Guide
  3. Using Machine Vision Tools
  4. Presence/Absence Tools
  5. Using Anomaly Detection
  6. Using Aurora Deep Learning Editor
  • Zebra Aurora Focus User Guide
Using Aurora Deep Learning Editor

Using Aurora Deep Learning Editor

Aurora Focus and Deep Learning Editor are directly connected and require compatible versions.
  1. Click to add new images to the editor, or click to add a new image and mark it as Good or Bad.
  2. Observe the
    Training Parameters
    to understand the complexity and downscaling of the model. Both contribute to the training time and the working time of the model.
    • Model Complexity - as density increases, the precision of the heatmap data visualization also increases, which in turn increases execution time.
    • Image Downscaling - each level reduces the width and height of the image by a factor of 2.
    Go to Using Pro Mode to understand how to use advanced parameters.
  3. Use the
    ROI
     tab to configure the region of interest. The size, shape, and location of the ROI can be changed to fit every image. Drawing tools help to mark the quiet zone inside the ROI. Fields outside the marked ROI are not calculated.
  4. After configuring the settings, click
    Start Training
    to start training the model.
If the Deep Learning Editor is being run for the first time, the first click launches the Deep Learning Service, and the second starts generating the model. The Deep Learning Service runs for approximately 30 minutes after disconnecting the device or stopping the emulator.
anomaly detect start model