3.4.5.1. Classification

_images/classification_alg.png — Classification

This tab allows for the classification of the Band set using the spectral signatures checked in ROI & Signature list. Several classification options are set in this tab which affect the classification process also during the Classification preview.

This tool allows for the selection of one the following algorithms:

Maximum Likelihood
Minimum Distance
Multi-layer Perceptron
Random Forest
Spectral Angle Mapping
Support Vector Machine

Also, it is possible to save and load a trained classifier.

Tip

Information about APIs of this tool in Remotior Sensus at this link .

3.4.5.1.1. Input 

Tool symbol and name	Description
Select input band set	select the input Band set to be classified
Use input normalization	if checked, normalize the input based on the selected method
Z-score	if checked with Use input normalized, Z-score normalization of input is performed
Linear scaling	if checked with Use input normalized, Linear scaling normalization of input is performed
Use training Macroclass ID	if checked, the classification is performed using
Use training Class ID the Macroclass ID (code MC ID of the signature)	if checked, the classification is performed using the Class ID (code C ID of the signature)

3.4.5.1.2. Algorithm 

This tool allows for the selection of the classification algorithm. The algorithm tab includes the available parameters.

3.4.5.1.2.1. Maximum Likelihood 

Use the Maximum Likelihood algorithm.

Tool symbol and name	Description
Use single threshold	if checked, it allows for the definition of a classification threshold (applied to all the spectral signatures); pixels are unclassified if probability is less than threshold value (max 100)
Signature threshold	if checked, thresholds Signature threshold are evaluated
	open the Signature threshold for the definition of signature thresholds
Save signature raster	if checked, in addition to the classification raster, for each spectral signature a raster is saved in the same output directory, which represents the distance between pixel and signature
Calculate classification confidence raster	if checked, calculate classification confidence raster

3.4.5.1.2.2. Minimum Distance 

_images/minimum_distance_alg.png — Minimum Distance

Use the Minimum Distance algorithm.

Tool symbol and name	Description
Use single threshold	if checked, it allows for the definition of a classification threshold (applied to all the spectral signatures); pixels are unclassified if distance is greater than threshold value
Signature threshold	if checked, thresholds Signature threshold are evaluated
	open the Signature threshold for the definition of signature thresholds
Save signature raster	if checked, in addition to the classification raster, for each spectral signature a raster is saved in the same output directory, which represents the distance between pixel and signature
Calculate classification confidence raster	if checked, calculate classification confidence raster

3.4.5.1.2.3. Multi-layer Perceptron 

_images/multi_layer_perceptron_alg.png — Multi-layer Perceptron

Use the Multi-Layer Perceptron algorithm.

Tool symbol and name	Description
Use framework scikit-learn	if checked, use scikit-learn framework (read this)
Use framework PyTorch	if checked, use PyTorch framework (read about this)
Hidden layer sizes	list of values separated by comma, where each value defines the number of neurons in a hidden layer (e.g.: 200, 100 for two hidden layers of 200 and 100 neurons respectively)
Max iter	set the maximum number of iterations
Activation	set the activation function (default: relu)
Alpha	set the weight decay (also L2 regularization term) for Adam optimizer
Training proportion	set the proportion of data to be used as training and the remaining part as test
Batch size	set the number of samples per batch for optimizer; if auto, the batch is the minimum value between 200 and the number of samples
Learning rate init	set initial learning rate
Cross validation	if checked, perform cross validation
Find best estimator with steps	if checked, find the best estimator iteratively with a number of steps
Calculate classification confidence raster	if checked, calculate classification confidence raster

Cross validation is a function provided by scikit-learn to avoid overfitting by splitting the training set into k smaller sets (read more . In particular, the function StratifiedKFold (with parameters n_splits=5, shuffle=True) is used to create 5 sets, each one containing approximately the same percentage of samples for each class as the complete set. This option can potentially increase significantly the computation time.

If Find best estimator with steps is checked, the algorithm tries to find the best estimator iteratively with the defined number of steps (the more the steps, the slower the process will be), by changing the algorithm parameters.

3.4.5.1.2.4. Random Forest 

_images/random_forest_alg.png — Random Forest

Use the Random Forest algorithm.

Tool symbol and name	Description
Number of trees	set the number of trees
Minimum number to split	set the minimum number of samples required to split an internal node
Max features	for node splitting, if empty all features are considered; if sqrt the square root of all the features, if integer number the number of features; if float number a fraction of all the features
One-Vs-Rest	if checked, perform One-Vs-Rest classification (read more)
Cross validation	if checked, perform cross validation
Balanced class weight	if checked, balanced weight is computed inversely proportional to class frequency
Find best estimator with steps	if checked, find the best estimator iteratively with a number of steps
Calculate classification confidence raster	if checked, calculate classification confidence raster

Cross validation is a function provided by scikit-learn to avoid overfitting by splitting the training set into k smaller sets (read more . In particular, the function StratifiedKFold (with parameters n_splits=5, shuffle=True) is used to create 5 sets, each one containing approximately the same percentage of samples for each class as the complete set. This option can potentially increase significantly the computation time.

If Find best estimator with steps is checked, the algorithm tries to find the best estimator iteratively with the defined number of steps (the more the steps, the slower the process will be), by changing the algorithm parameters.

If One-Vs-Rest is checked, the algorithm performs One-Vs-Rest classification which basically fits one classifier per class.

If Balanced class weight is checked, the algorithm gives all classes equal weight with a balanced weight that is computed inversely proportional to class frequency in the training data.

3.4.5.1.2.5. Spectral Angle Mapping 

_images/spectra_angle_mapping_alg.png — Spectral Angle Mapping

Use the Spectral Angle Mapping algorithm.

Tool symbol and name	Description
Use single threshold	if checked, it allows for the definition of a classification threshold (applied to all the spectral signatures); pixels are unclassified if spectral angle distance is greater than threshold value (max 90)
Signature threshold	if checked, thresholds Signature threshold are evaluated
	open the Signature threshold for the definition of signature thresholds
Save signature raster	if checked, in addition to the classification raster, for each spectral signature a raster is saved in the same output directory, which represents the distance between pixel and signature
Calculate classification confidence raster	if checked, calculate classification confidence raster

3.4.5.1.2.6. Support Vector Machine 

_images/support_vector_machine_alg.png — Support Vector Machine

Use the Support Vector Machine algorithm.

Tool symbol and name	Description
Regularization parameter C	set the regularization parameter C
Kernel	set the kernel (default: rbf)
Gamma	set the kernel coefficient gamma (default: scale)
Cross validation	if checked, perform cross validation
Balanced class weight	if checked, balanced weight is computed inversely proportional to class frequency
Find best estimator with steps	if checked, find the best estimator iteratively with a number of steps
Calculate classification confidence raster	if checked, calculate classification confidence raster

Cross validation is a function provided by scikit-learn to avoid overfitting by splitting the training set into k smaller sets (read more . In particular, the function StratifiedKFold (with parameters n_splits=5, shuffle=True) is used to create 5 sets, each one containing approximately the same percentage of samples for each class as the complete set. This option can potentially increase significantly the computation time.

If Find best estimator with steps is checked, the algorithm tries to find the best estimator iteratively with the defined number of steps (the more the steps, the slower the process will be), by changing the algorithm parameters.

If Balanced class weight is checked, the algorithm gives all classes equal weight with a balanced weight that is computed inversely proportional to class frequency in the training data.

3.4.5.1.3. Run 

It is possible to run the classification, or save and load a trained classifier.

Classification raster is a file .tif (a QGIS style file .qml is saved along with the classification); also other outputs can be optionally calculated. Outputs are loaded in QGIS after the calculation.

Tool symbol and name	Description
Load classifier	open an already save classifier file (.rsmo)
Save classifier	save the classifier to file (.rsmo), in order to be loaded later
RUN	run this function