Double-sided scanning in linux and converting to pdf. Dependency: sudo apt-get install xsane imagemagick

scan.sh

#!/bin/bash
scanimage --batch --batch-double
read -p "Flip papers in the feeder, last page on top. Press [Enter] key to start..."
endpage=$(echo "$(ls *.pnm -1 | wc -l) * 2" | bc)
scanimage --batch --batch-increment -2 --batch-start $endpage
for file in *.pnm; do convert $file $file.jpg; done
rm *.pnm
convert $(ls *.jpg -1v | paste) -compress jpeg -page A4 output.pdf 

Dependency
sudo apt-get install xsane imagemagick

Thanks, this is great!
You could also add ocrmypdf at the end for text recognition. Brings quite some more dependencies, tho (I think python + tesseract-ocr at the very least)

Awesome. The only nitpick is that if there are other jpg's in my current folder it will combine all of them into the final pdf. Also, it will rm all my preexisting pnm files. I've added a few small things

• temporary folder to make sure the work dir is clean.
• choose feeder from ADF instead of flatbed.
• manually set resolution and mode

#!/bin/bash
mkdir scan_tmp
cd scan_tmp
scanimage --batch --batch-double -d 'escl:http://192.168.100.142:80' --mode lineart  --resolution 300 --source ADF
read -p "Flip papers in the feeder, last page on top. Press [Enter] key to start..."
endpage=$(echo "$(ls *.pnm -1 | wc -l) * 2" | bc)
scanimage --batch --batch-increment -2 --batch-start $endpage -d 'escl:http://192.168.100.142:80' --mode lineart  --resolution 300 --source ADF
for file in *.pnm; do convert $file $file.jpg; done
# rm *.pnm
convert $(ls *.jpg -1v | paste) -compress jpeg -page A4 output.pdf
# rm *.jpg
cd ..
mv scan_tmp/output.pdf ./
rm -rf scan_tmp/

Lastly, my convert does not convert PDF by throwing out the following error:

convert-im6.q16: attempt to perform an operation not allowed by the security policy `PDF'

I followed the solution here to comment out <policy domain="coder" rights="none" pattern="PDF" /> inside /etc/ImageMagick-6/policy.xml.