set_generic_shape_model_object — Set iconic objects to the shape model.
set_generic_shape_model_object(Object : : ModelID, GenParamName : )
The operator set_generic_shape_model_object sets parameters
of the shape model ModelID based on the given Object.
The parameter GenParamName determines for which type of setting
the object is used. The following values are supported:
'high_noise_sample': Estimate and adapt the shape model
parameters to cope better with high noise images.
The parameter estimation is done based on the sample search image
Object. As a consequence the domain of Object should
contain the model and surrounding background showing noise that is typical
for search images.
The operator modifies the following parameters:
'pyramid_level_lowest'
'clutter_region': Set the clutter region for the shape model. If 'clutter_region' is set, the value of 'use_clutter' is automatically set to 'true'.
Note that the 'clutter_region' must not intersect with the final
model contours that are generated by train_generic_shape_model.
If the clutter region would need negative coordinates (e.g., clutter
regions for XLD models which have the center in the image origin), the
transformation has to be provided. To do so, use
set_generic_shape_model_param with 'clutter_hom_mat_2d'.
This operator modifies the state of the following input parameter:
During execution of this operator, access to the value of this parameter must be synchronized if it is used across multiple threads.
Object (input_object) (multichannel-)object(-array) → object
Iconic object to be set.
ModelID (input_control, state is modified) shape_model → (handle)
Handle of the shape model.
GenParamName (input_control) string-array → (string)
Parameter name.
Default value: 'high_noise_sample'
List of values: 'clutter_region', 'high_noise_sample'
If the parameters are valid, the operator
set_generic_shape_model_object returns the value 2 (H_MSG_TRUE).
If necessary an exception is raised.
train_generic_shape_model,
find_generic_shape_model
Matching