Skip to contents

Summarize Cluster Spectra with Optional Variance Propagation

Source: R/summarize_cluster_spectra.R
summarize_cluster_spectra.Rd

This is the recommended post-segmentation summary layer for Capivara. It keeps the segmentation step separate from downstream spectral products and can return median spectra, arithmetic mean spectra, summed spectra, and inverse-variance-weighted mean spectra.

Usage

summarize_cluster_spectra(
  cluster_result,
  var_cube = NULL,
  variance_inflation = 1
)

Arguments

cluster_result

A list returned by a segmentation function.

var_cube

Optional variance cube matching the dimensions of the original flux cube. It can be a FITS-like list with imDat or a raw 3-D array.

variance_inflation

Multiplicative factor applied to propagated variances. Use this to account for covariance if needed.

Value

A list with wavelength coordinates, cluster ids, cluster sizes, per-wavelength finite counts, median spectra, mean spectra, summed spectra, and, when var_cube is supplied, propagated sum variances and inverse-variance-weighted means.