The aim of B11 program is to transform and/or classify the data (image texture features) obtained from MaZda software. Andrzej Materka wrote its source code. The program was written in Object Pascal and was compiled using Delphi 6 Inprise package. It works under Microsoft Windows 9x/NT/2k/XP operating systems.
B11 program is used for two main tasks:
- Feature reduction through linear and nonlinear transformations
- Data visualization
- Data classification
B11 provides feature projection by converting the input data space to a new space, with reduced number of dimensions. Program displays both input and converted data in a form of a graph.
The following data processing procedures are implemented in B11:
- Principal Component Analysis (PCA)
- Linear Discriminant Analysis (LDA)
- Non-linear Discriminant Analysis (NDA)
B11 allows also feature vector classification, and evaluation of the usefulness of texture features calculated using MaZda to discrimination of different texture class - present in image regions. For data classification, the following classifiers are implemented:
- Nearest neighbour classifier (1-NN)
- Artificial neural network classifier (ANN)
Neural networks of the architecture defined during training can be tested using data sets composed offeature vectors calculated for images not used for the training.
<< Back | MaZda >>