Name
add_samples_image_class_gmmadd_samples_image_class_gmmAddSamplesImageClassGmmadd_samples_image_class_gmmAddSamplesImageClassGmmAddSamplesImageClassGmm — Add training samples from an image to the training data of a
Gaussian Mixture Model.
add_samples_image_class_gmmadd_samples_image_class_gmmAddSamplesImageClassGmmadd_samples_image_class_gmmAddSamplesImageClassGmmAddSamplesImageClassGmm adds training samples from the
ImageImageImageImageImageimage to the Gaussian Mixture Model (GMM) given by
GMMHandleGMMHandleGMMHandleGMMHandleGMMHandleGMMHandle. add_samples_image_class_gmmadd_samples_image_class_gmmAddSamplesImageClassGmmadd_samples_image_class_gmmAddSamplesImageClassGmmAddSamplesImageClassGmm is used to
store the training samples before a classifier to be used for the
pixel classification of multichannel images with
classify_image_class_gmmclassify_image_class_gmmClassifyImageClassGmmclassify_image_class_gmmClassifyImageClassGmmClassifyImageClassGmm is trained.
add_samples_image_class_gmmadd_samples_image_class_gmmAddSamplesImageClassGmmadd_samples_image_class_gmmAddSamplesImageClassGmmAddSamplesImageClassGmm works analogously to
add_sample_class_gmmadd_sample_class_gmmAddSampleClassGmmadd_sample_class_gmmAddSampleClassGmmAddSampleClassGmm. The ImageImageImageImageImageimage must have a number
of channels equal to NumDimNumDimNumDimNumDimNumDimnumDim, as specified with
create_class_gmmcreate_class_gmmCreateClassGmmcreate_class_gmmCreateClassGmmCreateClassGmm. The training regions for the
NumClassesNumClassesNumClassesNumClassesNumClassesnumClasses pixel classes are passed in
ClassRegionsClassRegionsClassRegionsClassRegionsClassRegionsclassRegions. Hence, ClassRegionsClassRegionsClassRegionsClassRegionsClassRegionsclassRegions must be a tuple
containing NumClassesNumClassesNumClassesNumClassesNumClassesnumClasses regions. The order of the regions
in ClassRegionsClassRegionsClassRegionsClassRegionsClassRegionsclassRegions determines the class of the pixels. If
there are no samples for a particular class in ImageImageImageImageImageimage an
empty region must be passed at the position of the class in
ClassRegionsClassRegionsClassRegionsClassRegionsClassRegionsclassRegions. With this mechanism it is possible to use
multiple images to add training samples for all relevant classes to
the GMM by calling add_samples_image_class_gmmadd_samples_image_class_gmmAddSamplesImageClassGmmadd_samples_image_class_gmmAddSamplesImageClassGmmAddSamplesImageClassGmm multiple
times with the different images and suitably chosen regions. The
regions in ClassRegionsClassRegionsClassRegionsClassRegionsClassRegionsclassRegions should contain representative
training samples for the respective classes. Hence, they need not
cover the entire image. The regions in ClassRegionsClassRegionsClassRegionsClassRegionsClassRegionsclassRegions should
not overlap each other, because this would lead to the fact that in
the training data the samples from the overlapping areas would be
assigned to multiple classes, which may lead to a lower
classification performance. Image data of integer type can be
particularly badly suited for modeling with a
GMM. RandomizeRandomizeRandomizeRandomizeRandomizerandomize can be used to overcome this problem, as
explained in add_sample_class_gmmadd_sample_class_gmmAddSampleClassGmmadd_sample_class_gmmAddSampleClassGmmAddSampleClassGmm.
- Multithreading type: reentrant (runs in parallel with non-exclusive operators).
- Multithreading scope: global (may be called from any thread).
- Processed without parallelization.
This operator modifies the state of the following input parameter:
The value of this parameter may not be shared across multiple threads without external synchronization.
Regions of the classes to be trained.
Standard deviation of the Gaussian noise added
to the training data.
Default value: 0.0
Suggested values: 0.0, 1.5, 2.0
Restriction: Randomize >= 0.0
If the parameters are valid, the operator
add_samples_image_class_gmmadd_samples_image_class_gmmAddSamplesImageClassGmmadd_samples_image_class_gmmAddSamplesImageClassGmmAddSamplesImageClassGmm returns the value 2 (H_MSG_TRUE). If
necessary an exception is raised.
create_class_gmmcreate_class_gmmCreateClassGmmcreate_class_gmmCreateClassGmmCreateClassGmm
train_class_gmmtrain_class_gmmTrainClassGmmtrain_class_gmmTrainClassGmmTrainClassGmm,
write_samples_class_gmmwrite_samples_class_gmmWriteSamplesClassGmmwrite_samples_class_gmmWriteSamplesClassGmmWriteSamplesClassGmm
read_samples_class_gmmread_samples_class_gmmReadSamplesClassGmmread_samples_class_gmmReadSamplesClassGmmReadSamplesClassGmm
classify_image_class_gmmclassify_image_class_gmmClassifyImageClassGmmclassify_image_class_gmmClassifyImageClassGmmClassifyImageClassGmm,
add_sample_class_gmmadd_sample_class_gmmAddSampleClassGmmadd_sample_class_gmmAddSampleClassGmmAddSampleClassGmm,
clear_samples_class_gmmclear_samples_class_gmmClearSamplesClassGmmclear_samples_class_gmmClearSamplesClassGmmClearSamplesClassGmm,
get_sample_num_class_gmmget_sample_num_class_gmmGetSampleNumClassGmmget_sample_num_class_gmmGetSampleNumClassGmmGetSampleNumClassGmm,
get_sample_class_gmmget_sample_class_gmmGetSampleClassGmmget_sample_class_gmmGetSampleClassGmmGetSampleClassGmm
Foundation