bivariate colour palette

gistfile1.txt

# Code by Dave Redding


library(raster)
zzz=100
colsX<-raster(nrow=zzz,ncol=zzz)
values(colsX)<-1:ncell(colsX)
extent(colsX)<-c(-100,100,-100,100)

cols<-c("#00FF00","#FFFFFF")
#cols<-c("#00FF00","#000000")
colX1<-colorRampPalette(cols,space="rgb",interpolate ="spline",bias=1.5)
c1<-colX1(zzz)

cols2<-c("#FC0000","#0000FC")
colX2<-colorRampPalette(cols2,space="rgb",interpolate ="spline",bias=1.5)
c2<-colX2(zzz)

for (i in 1:100){
cols3<-c(c1[i],c2[i])
colX3<-colorRampPalette(cols3,space="rgb",interpolate ="spline")
c3<-colX3(zzz)
if(i==1){c4<-c3} else{c4<-c(c4,c3)}
}

plot(colsX,col=c4,legend=F)

res1<-matrix(nrow=zzz,ncol=zzz,data=c4, byrow = TRUE)


#'@example
#' data <- data.frame(x = rnorm(100), y = rnorm(100), z1 = runif(100), z2 = runif(100))
#' data$z1percentile <- as.numeric(cut(data$z1, 100))
#' data$z2percentile <- as.numeric(cut(data$z2, 100))
#' data$col <- sapply(1:nrow(data), function(x) res1[data$z1percentile[x], data$z2percentile[x]])
#  layout(matrix(c(1,2), 1, 2, byrow = TRUE), widths = c(0.7, 0.3))
#' plot(y ~ x, data, col = data$col)
#' image(as.matrix(colsX),col=c4,legend=F)