defuzzify: Defuzzify fuzzy classification

In rcaiman: CAnopy IMage ANalysis

Defuzzify fuzzy classification

Description

This function translates degree of membership into Boolean logic using a regional approach. The result will ensure that the fuzzy and Boolean version will agree at the chosen level of aggregation (controlled by the argument segmentation). This method makes perfect sense to translate a subpixel classification of gap fraction–or a linear ratio \insertCiteLang2013rcaiman–into a binary product.

Usage

defuzzify(mem, segmentation)

Arguments

mem	An object of the class SpatRaster. Degree of membership.
segmentation	An object of the class SpatRaster, such as the result of a call to sky_grid_segmentation.

Details

This method is also available in the HSP software package \insertCiteLang2013rcaiman.

Value

An object of the class SpatRaster containing binary information.

References

\insertAllCited

See Also

Other Tool Functions: colorfulness(), extract_dn(), extract_feature(), extract_rl(), extract_sky_points(), masking(), read_bin(), read_caim(), write_bin(), write_caim()

Examples

## Not run: 
path <- system.file("external/DSCN4500.JPG", package = "rcaiman")
caim <- read_caim(path, c(1280, 960) - 745, 745 * 2, 745 * 2)
z <- zenith_image(ncol(caim), lens("Nikon_FCE9"))
a <- azimuth_image(z)
r <- gbc(caim$Blue)
r[is.na(z)] <- 0 # because FOV > 180
bin <- ootb_mblt(r, z, a)
plot(bin$bin)
ratio <- r / bin$sky_s
ratio <- normalize(ratio, 0, 1, TRUE)
plot(ratio)
g <- sky_grid_segmentation(z, a, 10)
bin2 <- defuzzify(ratio, g)
plot(bin2)
plot(bin$bin - bin2)

## End(Not run)

rcaiman documentation built on May 31, 2023, 7:35 p.m.