Use of this function requires a license for Whitebox Workflows for Python Professional (WbW-Pro). Please visit www.whiteboxgeo.com to purchase a license.
This tool takes a two or more component images (inputs
), and the principal component analysis (PCA) report derived using the principal_component_analysis tool, and performs the inverse PCA transform to derive the original series of input images. This inverse transform is frequently performed to reduce noise within a multi-spectral image data set. With a typical PCA transform, high-frequency noise will commonly map onto the higher component images. By excluding one or more higher-valued component images from the input component list, the inverse transform can produce a set of images in the original coordinate system that exclude the information contained within component images excluded from the input list. Note that the number of output images will also equal the number of original images input to the principal_component_analysis tool. The output images will be named automatically with a "inv_PCA_image" suffix.
def inverse_pca(self, rasters: List[Raster], pca_report_file: str) -> List[Raster]: ...