apply_FCM_m: Apply the improved Fuzzy Cluster Method (FCMm) to new data...

In bishun945/FCMm: Fuzzy Cluster Method Based on the Optimized m Value (Fuzzifier)

Description

Application of the improved Fuzzy Cluster Method (FCMm) for new Rrs data based on default cluster settings or user-defined clusters (trained by FCM.new).

Usage

apply_FCM_m(
  Rrs,
  wavelength = NULL,
  Rrs_clusters = NULL,
  m_used = 1.36,
  do.stand = FALSE,
  stand = NULL,
  default.cluster = TRUE,
  quality_check = FALSE,
  option.plot = FALSE,
  color_palette = NULL
)

Arguments

Rrs	Data.frame, the input Rrs of FCM.
wavelength	Numeric vector, used for applying FCM. Default use the data from Bi_clusters.rda
Rrs_clusters	Data.frame, used for applying FCM. Default use the data from Bi_clusters.rda
m_used	Number, Used fuzzifier value
do.stand	Logical, whether to normalized the Rrs data (both for Input and Centroids). Default as FALSE means do not.
stand	Deprecated; same to do.stand.
default.cluster	Logical, whether to use the default clusters. Default use the data from Bi_clusters.rda
quality_check	Logical, quality chech option (default as FALSE)
option.plot	Logical, whether to plot the result. Default as FALSE
color_palette	The palette of cluster color. Default as RdYlBu(res$K) if NULL. It could be RdYlBu(k), Spectral(k), HUE(k) or other color values with same length of cluster number. k means the cluster number.

Value

A list including several results of function apply_FCM_m()

• x The raw input Rrs dataframe with unit sr^-1

• x.stand The standardized Rrs dataframe, if stand=F

• d Distance to each cluster

• u Membership values

• Area Spectral intergration of each sample

• cluster Defined by the maximum of membership

• quality The quality of the cluster results.

• m.used The used value of fuzzifier(m)

• K Cluster number

• p.group A ggplot list for plotting the cluster result

• p.group.facet Recommened! p.group with facet to see each cluster results more clearly

• dt.melt Dataframe used for ggplot including dt_plot_x, dt_plot_v, and cp.

References

Bi S, Li Y, Xu J, et al. Optical classification of inland waters based on an improved Fuzzy C-Means method[J]. Optics Express, 2019, 27(24): 34838-34856.

See Also

Other Fuzzy cluster functions: FCM.new(), FuzzifierDetermination(), apply_to_image(), cal_memb(), generate_param(), plot_spec_from_df()

Examples

library(FCMm)
data("WaterSpec35")
data("Bi_clusters")
Rrs <- WaterSpec35[,3:17]
result <- apply_FCM_m(Rrs=Rrs, option.plot=TRUE)

bishun945/FCMm documentation built on Oct. 15, 2021, 6:43 p.m.