I am trying to place images (png) on top of a map using the set pixmap
command.
It does work but the transparency of the png is not getting "displayed", and I get a white background instead.
Can the pixmap
feature handle the transparency property of an image?
Here is the script:
$DATA<<EOD
1,"Washington",630.2000,1591.1800
EOD
$DATA01<<EOD
1 "C:\\Temp\\Images\\colors.png" 630.2000 1591.1800
EOD
set datafile separator comma
set style data point
set term windows enhanced size 3000.00,2000.00
unset xtics; unset ytics; unset x2tics; unset y2tics
set grid noxtics
set grid noytics
set border 0
set colorsequence classic
set key off
set for [i=1:1] pixmap i word($DATA01[i], 2) center at word($DATA01[i], 3), word($DATA01[i], 4) width 100 height 100
plot 'C:\Magic\XPA\46s\Add_On\ChartGenerator\Geography\USA_Curved_Names_Blue.png.jpg' binary filetype=JPG w rgbimage,\
$DATA every ::0::0 using ($3):($4) notitle,\
$DATA using ($3):($4):($2) with labels font "Tahoma,20" textcolor rgb "purple" center offset 0,1 notitle
Roberto
I do not know what is going on here exactly, but it has something to do with the image file itself. If I open that file in GIMP and export it to a new file with no intervening edit/modify operations, then the exported file works correctly as a pixmap with transparent areas.
Comparing the output of identify -verbose original.png
and identify -verbose redo.png
doesn't show anything that stands out to me except possibly the png:IHDR.color_type
entries.
I attach the re-exported file here - try it in place of your original image file and see if that works.
For what it's worth, here is the diff between identify
run on the two files:
diff -urp original.identify redo.identify
--- original.identify 2023-10-25 22:38:42.002322272 -0700
+++ redo.identify 2023-10-25 22:38:57.165582322 -0700
@@ -1,5 +1,5 @@
Image:
- Filename: original.png
+ Filename: redo.png
Permissions: rw-r--r--
Format: PNG (Portable Network Graphics)
Mime type: image/png
@@ -703,25 +703,26 @@ Image:
Compression: Zip
Orientation: Undefined
Properties:
- date:create: 2023-10-26T02:58:38+00:00
- date:modify: 2023-10-26T02:58:34+00:00
- date:timestamp: 2023-10-26T06:23:13+00:00
+ date:create: 2023-10-26T06:22:24+00:00
+ date:modify: 2023-10-26T06:22:24+00:00
+ date:timestamp: 2023-10-26T06:23:23+00:00
+ png:bKGD: chunk was found (see Background color, above)
png:IHDR.bit-depth-orig: 8
png:IHDR.bit_depth: 8
- png:IHDR.color-type-orig: 2
- png:IHDR.color_type: 2 (Truecolor)
+ png:IHDR.color-type-orig: 6
+ png:IHDR.color_type: 6 (RGBA)
png:IHDR.interlace_method: 0 (Not interlaced)
png:IHDR.width,height: 512, 512
png:pHYs: x_res=3780, y_res=3780, units=1
- png:tRNS: chunk was found
+ png:tIME: 2023-10-26T06:22:24Z
signature: cb4667f54b948708bd7fd893e4668ca843bf816392a4bad8c8917b967cf1f1ea
Artifacts:
verbose: true
Tainted: False
- Filesize: 11772B
+ Filesize: 11446B
Number pixels: 262144
Pixel cache type: Memory
- Pixels per second: 21.2189MP
+ Pixels per second: 21.0638MP
User time: 0.010u
Elapsed time: 0:01.012
Version: ImageMagick 7.1.1-11 Q16-HDRI x86_64 21206 https://imagemagick.org
If you spot anything relevant there, let me know. If it's a reproducible file property that can be tested for, maybe gnuplot can be taught to fix it or at least emit a warning.