Is it possible to get fine anti-aliasing for exported ggplot plots? I have tried the Cairo
package as well as few different devices, but they all seems to have jagged edges.
library(ggplot2)
library(Cairo)
p <- ggplot(data.frame(x=1:5,y=1:5),aes(x=x,y=y))+
geom_text(aes(2.5,2.5),label="Brown Fox bla bla..",size=5)+
labs(x=NULL,y=NULL)+
theme_bw()+
theme(plot.background=element_blank(),
plot.margin = margin(c(0,0,0,0)),
axis.title = element_blank(),
axis.text = element_blank(),
axis.ticks = element_blank(),
axis.line = element_blank(),
panel.grid = element_blank(),
panel.border = element_blank(),
panel.background = element_blank())
png("test-nocairo.png",height=2,width=6,units="cm",res=300)
print(p)
dev.off()
png("test-cairo.png",height=2,width=6,units="cm",res=300,type="cairo")
print(p)
dev.off()
tiff("test-cairo.tiff",height=2,width=6,units="cm",res=300,type="cairo")
print(p)
dev.off()
ggsave("test-ggsave.png",height=2,width=6,units="cm",dpi=300,type="cairo")
PNG no cairo
PNG cairo
TIFF cairo
PNG ggsave cairo
For my purpose, it is important that the images are PNG or TIFF (lossless) at 300dpi. I am aware that I can export to a vector format (SVG,PDF etc) and then convert to PNG/TIFF using another program, but that is obviously extra work. I am curious if there are any solutions in R that I am overlooking.
For reference, above is the quality of rendering from photoshop. PNG Arial 14pt.
Ok. I may have stumbled upon something here. When I use annotate
rather than geom_text
, the cairo
anti-aliasing seems to work.
p <- ggplot(data.frame(x=1:5,y=1:5),aes(x=x,y=y))+
annotate("text",x=2.5,y=2.5,label="Brown Fox bla bla..",size=5)+
labs(x=NULL,y=NULL)+
theme_bw()+
theme(plot.background=element_blank(),
plot.margin = margin(c(0,0,0,0)),
axis.title = element_blank(),
axis.text = element_blank(),
axis.ticks = element_blank(),
axis.line = element_blank(),
panel.grid = element_blank(),
panel.border = element_blank(),
panel.background = element_blank())
png("test-annotate-cairo.png",height=2,width=6,units="cm",res=300,type="cairo")
print(p)
dev.off()
So, it seems like geom_text
is overplotting the same text which might be the issue. I thought this overplotting thing was fixed at some point. I think there is still room for improvement with the anti-aliasing, but this is much better than before.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With