ZEISS Microscopy Knowledge Base

Help

ZEISS Microscopy Knowledge Base

Intellesis Object Classification

This module offers the functionality to classify objects based on measured parameters of an analyzed image using machine-learning algorithms and to create and to train such a model for object classification. Since the input for the object classification is an analyzed image containing a result table, the functionality of the Image Analysis module is also required for the complete workflow.

Licensing and Functionalities of Intellesis Object Classification

Some functionality of Intellesis Object Classification is generally available, but the full functionality requires the AI Toolkit license.

Basic functionality

The general available functionality includes:

Importing and exporting models.
Managing the models, including renaming and deleting.
Running a model with the Intellesis Object Classification workbench and tool or in OAD (if you have licensed the Developer Toolkit).

Licensed functionality

If you have licensed this functionality and activated it under Maintenance > Toolkit Manager, the following additional functionality is available:

Creating and training a new model.
Retraining an existing model.

Workflow Overview for Intellesis Object Classification

The workflow to train and use an Intellesis object classification model consists of three main steps:

Analyzing an image, see Image Analysis.
Training an object classification model, see Creating and Training an Intellesis Object Classification Model.
Using the trained model to classify objects in analyzed images, see Using an Intellesis Object Classification Model for Image Processing.

Creating and Training an Intellesis Object Classification Model

You have licensed the AI Toolkit and activated it under Maintenance > Toolkit Manager.
You are in Manage Templates.
You have a CZI image with image analysis results. For detailed information on analysis, see Image Analysis.

For Show, select Intellesis Object Classification Models.
In the top right tool bar, click .
The user interface for training opens, see Training User Interface Intellesis Object Classification.
In the top left corner, click Import Images. To open an image from the archive, click Import from Archive.
A file browser or the archive browser opens.
Select the image for training from the file system and click Open. When loading the image from the archive, select the image and click OK.
If the image analysis contains more than one class and channel, the Select Region Class and Channel(s) dialog opens.
Select the region class and the channels you want to use for classification and click OK.
The image is displayed in the list. Note that all imported images are included in your model.
If you have imported an image from the archive, it is copied to the location defined under Maintenance > General Options > Intellesis.
Select the image from the list.
The image is displayed, and the table shows the data of the analysis result.
On the left side, add as many classes as you need for your object classification by clicking Add Class.
You have created the classes that you want to distinguish.
To change the label color for a class, click on the color field of the list entry and select one from the window.
To rename a class, right click on the entry, select Rename, enter a new name and click . Alternatively, double click the name entry, enter a new one and click , or press Enter.
In the Labeling Options tab, click .
You are now in labeling mode.
In the classes list, select a class and click on an object that belongs to this class in the image or in the table.
You have labeled the object and assigned it a class.
Repeat the labeling for the objects of the different classes you created.
Click Train & Classify.
Your model is trained based on the labeling and a prediction is displayed.
If you are not satisfied with the result, you can label more objects, click Train & Classify again, and iteratively improve your results.
When you are satisfied with the result, click Finish.
All changes are saved, and the training interface is closed.

You have successfully created and trained a model for object classification. You can now use it to classify objects in your analyzed images with the Intellesis Object Classification tool, see Using an Intellesis Object Classification Model for Image Processing.

Importing an Intellesis Object Classification Model

You have a trained model available for import.

Under Manage Templates > Show, select Intellesis Object Classification Models.
In the top right tool bar, click .
A file browser opens.
In the file browser, select the model file from the file system, and click Import.

You have imported a model, and it is available in the list of object classification models. You can now use it to classify objects in your images or open it (e.g. for training) by double clicking on it.

Exporting an Intellesis Object Classification Model

NOTICE

Blue circular sign with white exclamation mark

Risk of non-functioning models in new ZEN core versions

In case of changes in Python libraries, trained models can stop working on a new version of ZEN core and need to be retrained first. Retraining is only possible if the model contains the images, or the images are generally still available. To be able to retrain your model, consider the following solutions:

Export your model with your images, see Exporting an Intellesis Object Classification Model with Images.
Make a backup of the images used for training the model, e.g. on your (external) hard drive.

You are in Manage Templates.
You have created an object classification model, see Creating and Training an Intellesis Object Classification Model.

For Show, select Intellesis Object Classification Models.
Right-click the model you want to export and select Export.
A file browser opens.
Select the file location and click OK.

The model is exported to that location. It contains the trained classification routine. Such a model is meant to be used only for classifying objects in an analyzed image, but not for the training process anymore.

Exporting an Intellesis Object Classification Model with Images

You are in Manage Templates.
You have created an object classification model, see Creating and Training an Intellesis Object Classification Model.

For Show, select Intellesis Object Classification Models.
Right-click the model you want to export and select Export (with Images).
A file browser opens.
Select the file location and click OK.

The model is exported to that location. It contains the trained object classification routine and the images used for the training. Such a model can be used for classifying objects in an analyzed image and can be retrained in the future.

Using an Intellesis Object Classification Model for Image Processing

You have created and trained an object classification model, see Creating and Training an Intellesis Object Classification Model.
You are in Free Mode or Job Mode and have added the Intellesis Object Classification tool.
You have opened the analyzed image(s) for which you want to use the classification model.

In the Intellesis Object Classification tool, click .
The Open Template dialog opens.
Select the Model you want to use for object classification and click OK.
If you want to display all statistical features in the result table (and not only the features that were selected in the image analysis), activate Append Classification Features to Result Table.
Click Apply.
The objects in your image are now classified based on the trained model.

The classified image is displayed, and the table is updated, including the label for each object and the confidence.
You can now extract a result table (including classification result and analysis features) with the Create Result Table tool and subsequently create a table with only your classification result using the Category Histogram tool for table processing.