Once you have trained a model for segmentation, you can also use it in the Image Analysis for further analysis. To use the trained model, you must first create a new image analysis (IA) setting (*.CZIAS format) first. To use the trained model there are two options:

  • Create a new image analysis setting from the Intellesis segmentation model. This is the only option for Intellesis segmentation models trained on a multi-channel image.
  • Use an Intellesis segmentation model in the Automatic Segmentation step of an image analysis setting, see Automatic Segmentation. This option is only possible for models trained on single-channel images, but allows you to create image analysis settings with a complex hierarchy.
  1. You are in the Intellesis Segmentation tool.
  1. Select the trained model which you want to use to create an analysis setting, click , and select Create Analysis Setting.
  2. The dialog for saving the setting opens. The setting will be saved as *.czias file in the ZEN default folder for image analysis settings (usually under User/Documents/Carl Zeiss/ZEN/Documents/Image Analysis Settings).
  3. Click Save.
  4. The file is saved.
  5. Now change to the Image Analysis tool and select the setting from the dropdown list. Note that the setting is only available in the dropdown list when you have used the default folder for saving. Otherwise the setting must be loaded from the file system (specific location) via the Import option.
  6. The image analysis setting is loaded with the classes defined in the Intellesis segmentation model.
  7. You can now continue with setting up an image analysis. For more information, see Image Analysis Wizard.
  8. The image displays a sandstone dataset segmented using Intellesis Segmentation inside the Image Analysis Wizard and showing the actual segmentation step. Instead of conventional thresholds, the Intellesis segmentation model is used to segment the image.
  9. With the Min. Confidence (%) parameter, it is possible to exclude pixels where the model exhibits a low confidence level, applicable to all classes.
  10. The binary functions Fill Holes and Separate are applied solely to the binary masks produced by the Intellesis segmentation step, making them independent of the actual segmentation process.
  11. Sandstone Dataset segmented using Intellesis Segmentation in the Image Analysis Wizard showing the measurement results for one particular class (shown in green).