When I work with fine-scale movement data I inevitably want to make movies of them and share with collaborators. Many iterations later, I have long since lost my code (and the data) that I originally used to make these gifs but am too lazy to regenerate.
For example these two gifs show the trajectory of a pregnant moose and her step length (when the trajectory turns red, the moose is giving birth)
I need to sew these two gifs together, and below is how I figured out how to do it. This answer from stack overflow got me going in the right direction, but I modified the script a bit.
Will need a total of three folders, one for each of the gifs plus one for the sewn version. Put the original gifs in the first and second folder.
mkdir first
mv moose_step.gif first/
mkdir second
mv moose_step-2.gif second/
mkdir third
cd third/
Unpack both gifs and rename based on image number
cd first/
convert moose_step_simple.gif x%04d.gif
cd ..
cd second/
convert moose_step-2.gif x%04d.gif
cd ..
Navigate to third folder, this is where the concatenated gif will end up. I’m using the montage
function which puts two images together and looping across all files. The -geometry
argument is important as it determines how the images fit together.
In this case moose_step_simple.gif
is a 800x800 px image, and moose_step-2.gif
is an 800x200 px image and I want stack moose_step_simple.gif
above the moose_step-2.gif
.
cd third/
for filename in ../first/*
do
filename=`basename $filename`
montage -tile 1x2 -geometry 800x800 ../first/$filename ../second/$filename concat$filename
done
My hack was to change the original code to -tile 1x2
which I’m guessing means one row, two columns. I changed the geometry argument to the largest image (800x800) which means there’s some awkward whitespace around the 800x200 image. But because the backgrounds both white it doesn’t matter.
I should note: running this gives me the error “montage: unable to read font (null) @ error/annotate.c/RenderFreetype/1152.” which I ignore. I think it has something to do with the correct fonts installed (as per here). But since neither of these images have font, I can get away without having to worry about it.
Finally to make a new output, put together the images, also optimize and remove background to make the final gif smaller (goes from 18 Mb to 1.2 Mb roughly).
convert -layers optimize -background None concatx* output.gif
rm -rf ../first
rm -rf ../second
rm concat*
Volia!