License Information

Use of this function requires a license for Whitebox Workflows for Python Professional (WbW-Pro). Please visit www.whiteboxgeo.com to purchase a license.

Description

This tool performs a supervised parallelepiped classification using training site polygons (polys) and multi-spectral images (inputs). This classification method uses the minimum and maximum reflectance values for each class within the training data to characterize a set of parallelepipeds, i.e. multi-dimensional geometric shapes. The algorithm then assigns each unknown pixel in the image data set to the first class for which the pixel's spectral vector is contained within the corresponding class parallelepiped. Pixels with spectral vectors that are not contained within any class parallelepiped will not be assigned a class in the output image.

Like all supervised classification methods, this technique relies heavily on proper selection of training data. Training sites are exemplar areas of known and representative land cover type. The algorithm determines the spectral signature of the pixels within each training area, and uses this information to define the mean vector of each class. It is preferable that training sites are based on either field-collected data or fine-resolution reference imagery. In selecting training sites, care should be taken to ensure that they cover the full range of variability within each class. Otherwise the classification accuracy will be impacted. If possible, multiple training sites should be selected for each class. It is also advisable to avoid areas near the edges of land-cover patches, where mixed pixels may impact the purity of training site reflectance values.

After selecting training sites, the reflectance values of each land-cover type can be assessed using the evaluate_training_sites tool. In particular, the distribution of reflectance values should ideally be non-overlapping in at least one band of the multi-spectral data set.

See Also

evaluate_training_sites, min_dist_classification

Function Signature

def parallelepiped_classification(self, input_rasters: List[Raster], training_data: Vector, class_field_name: str) -> Raster: ...

Project Links

WbW Homepage User Manual Support WbW