Adopt new parameters that are used to create the model components into the training result.

With cluster_model_components you can modify parameters after a first training has been performed using train_model_components. cluster_model_components sets the criterion AmbiguityCriterion that is used to solve the ambiguities, the maximum contour overlap MaxContourOverlap, and the cluster threshold of the training result ComponentTrainingID to the specified values. A detailed description of these parameters can be found in the documentation of train_model_components. By modifying these parameters, the way in which the initial components are merged into rigid model components changes. For example, the greater the cluster threshold is chosen, the fewer initial components are merged.

The rigid model components are returned in ModelComponents. In order to receive reasonable results, it is essential that the same training images that were used to perform the training with train_model_components are passed in TrainingImages. The pose of the newly clustered components within the training images is determined using the shape-based matching. As in train_model_components, one can decide whether the shape models should be pregenerated by using set_system('pregenerate_shape_models',...).

Thus, you can select suitable parameter values interactively by repeatedly calling inspect_clustered_components with different parameter values and then setting the chosen values by using get_training_components.

* Get the model image.
read_image (ModelImage, 'model_image.tif')
* Define the regions for the initial components.
gen_rectangle2 (InitialComponentRegions, 212, 233, 0.62, 167, 29)
gen_rectangle2 (Rectangle2, 298, 363, 1.17, 162, 34)
gen_rectangle2 (Rectangle3, 63, 444, -0.26, 50, 27)
gen_rectangle2 (Rectangle4, 120, 473, 0, 33, 20)
InitialComponentRegions := [InitialComponentRegions,Rectangle2]
InitialComponentRegions := [InitialComponentRegions,Rectangle3]
InitialComponentRegions := [InitialComponentRegions,Rectangle4]
* Get the training images
TrainingImages := []
for i := 1 to 4 by 1
    read_image (TrainingImage, 'training_image-'+i$'02'+'.tif')
    TrainingImages := [TrainingImages,TrainingImage]
endfor
* Extract the model components and train the relations.
train_model_components (ModelImage, InitialComponentRegions, TrainingImages,
                        ModelComponents, 22, 60, 30, 0.65, 0, 0, rad(60),
                        'speed', 'rigidity', 0.2, 0.5, ComponentTrainingID)
* Find the best value for the parameter ClusterThreshold.
inspect_clustered_components (ModelComponents, ComponentTrainingID,
                              'rigidity', 0.2, 0.4)
* Adopt the ClusterThreshold into the training result.
cluster_model_components (TrainingImages, ModelComponents,
                          ComponentTrainingID, 'rigidity', 0.2, 0.4)
* Create the component model based on the training result.
create_trained_component_model (ComponentTrainingID, -rad(30), rad(60), 10,
                                0.5, 'auto', 'auto', 'none', 'use_polarity',
                                'false', ComponentModelID, RootRanking)

If the parameter values are correct, the operator cluster_model_components returns the value 2 (H_MSG_TRUE). If the input is empty (no input images are available) the behavior can be set via set_system('no_object_result',<Result>). If necessary, an exception is raised.

cluster_model_components is processed completely exclusively without parallelization.

train_model_components, inspect_clustered_components

get_training_components, create_trained_component_model, modify_component_relations, write_training_components, get_component_relations, clear_training_components, clear_all_training_components

Matching