Commit 67b19129 authored by Cédric Traizet's avatar Cédric Traizet
Browse files

comments added/modified

parent 2c525d93
This module contains a new dimensionality reduction framework for the Orfeo Toolbox.
The framework is based on Machine learning models and a dimensionality reduction algorithm
can be trained and used using the model class, in the same fashion as Machine Learning Models
for Classification (supervised or unspervised) and Regression.
The algorithms featured in the module are (27/06/2017) :
- autoencoders and multi layer autoencoders, with several regularization options
Autoencoders and PCA models are using Shark ML library, while SOM model is based on the SOM classes of the OTB.
More specifically, the module contains :
- Autoencoder models, PCA models and SOM models, with methods for training, serialization and prediction (i.e. reduction)
- A Dimensionality Reduction Model Factory and a factories for each model, which allow the user to create objects of the model classes
- A (OTB ImageToImage) filter that can be used to perform dimensionality reduction on an image. This filter supports threading and streaming
- An application for training the models according to a shapefile
- An application for using a trained model on a shapefile
- An application for using a trained model on an image (using the filter)
/!\ Work In Progress /!\
......@@ -112,33 +112,27 @@ protected:
void DoInit() ITK_OVERRIDE
SetDescription("Performs a prediction of the input image according to a regression model file.");
SetDescription("Performs dimensionality reduction of the input image according to a dimensionality reduction model file.");
// Documentation
SetDocName("Predict Regression");
SetDocLongDescription("This application predict output values from an input"
" image, based on a regression model file produced by"
" the TrainRegression application. Pixels of the "
"output image will contain the predicted values from"
"the regression model (single band). The input pixels"
SetDocLongDescription("This application reduces the dimension of an input"
" image, based on a dimensionality reduction model file produced by"
" the DimensionalityReductionTrainer application. Pixels of the "
"output image will contain the reduced values from"
"the model. The input pixels"
" can be optionally centered and reduced according "
"to the statistics file produced by the "
"ComputeImagesStatistics application. An optional "
"input mask can be provided, in which case only "
"input image pixels whose corresponding mask value "
"is greater than 0 will be processed. The remaining"
" of pixels will be given the value 0 in the output"
" image.");
"ComputeImagesStatistics application. ");
SetDocLimitations("The input image must contain the feature bands used for"
" the model training (without the predicted value). "
" the model training. "
"If a statistics file was used during training by the "
"TrainRegression, it is mandatory to use the same "
"statistics file for prediction. If an input mask is "
"used, its size must match the input image size.");
"statistics file for prediction.");
SetDocSeeAlso("TrainRegression, ComputeImagesStatistics");
SetDocSeeAlso("DimensionalityReductionTrainer, ComputeImagesStatistics");
......@@ -187,7 +181,7 @@ private:
unsigned int nbFeatures = inImage->GetNumberOfComponentsPerPixel();
// Load svm model
// Load DR model using a factory
otbAppLogINFO("Loading model");
m_Model = DimensionalityReductionModelFactoryType::CreateDimensionalityReductionModel(GetParameterString("model"),
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment