2.4.5 Learning

Supervised classification

Supervised classification is a procedure in which individual items are placed into groups based on quantitative information on one or more characteristics inherent in the items and based on a training set of previously labeled items.

The supervised classification module is based on the Support Vector Machine method which consists in searching for the separating surface between 2 classes by the determination of the subset of training samples which best describes the boundary between the 2 classes. This method can be extended to be able to classify more than 2 classes.

The module allows to interactivelly describe learnings samples which corresponds to polygons samples on the input images.

Then a SVM model is derived from this learning sample which allows to classify each pixel of the input image in one of the defined class.

Non-supervised classification

The non supervised classification module is based on the Kmeans algorithm. The GUI allows to modify parameters of the algorithm and produce a label image.