HowTo: Using Scanbot and Nextcloud for scanning, OCR and file management


#1

(Originally from : https://medium.com/@mathiasconradt/using-scanbot-and-nextcloud-for-scanning-ocr-and-file-management-229750420882)

For my personal documents, I use a Nextcloud installation on a RaspberryPi 3 and reachable from the outside. In order to digitalize paper documents that I receive, I’ve been using Scanbot for some years now.

It’s an Android app that lets you photograph documents and save them as images or (my preference) pdf files.

With Scanbot Pro you can even let an OCR engine analyze the text, which stores the text as an extra layer in the pdf, making it searchable and even indexable (great if you use the Nextcloud Nextant app powered by Solr).

In order to connect Scanbot with your Nextcloud, you can simply define a custom WebDav server endpoint and select a folder in the app preferences. In Scanbot, you can then upload the scanned documents manually or have them uploaded automatically.

(1)

(2)

(3)

In my case, I have a folder named “Scanbot” in my root directory under my user. So the WebDav endpoint url would look like this:

(4)

https://<mydomain>/remote.php/dav/files/<username>/Scanbot/

(5)

When I now scan new documents in the Scanbot app, they are automatically uploaded to my “Scanbot” folder on my Nextcloud.


#2

Thanks @mathiasconradt for posting!

Community! I have a spare Scanbot Pro license! Anyone interested in getting it, give the above OP a :heart: and on Friday 4th I’ll randomly select a winner!


#3

Scanbot also has an option for an Owncloud account (which is compatible with Nextcloud) to make the setup a bit easier because you only need your Nextcloud domain instead of the full Webdav address.


#4

@mathiasconradt great piece of work!! <3 <3 <3

only one little concern… wouldn’t nextant go out of service sometime during next future (like from NC 14 on)? have you tried replacing it with Full Text Search as well?

apart from that… wonderful. thanks so much for your contribution


#5

For whatever reason, I was not able to use the ownCloud connector for my instance, so also went with webDAV. Didn’t need a full path either as it brought up an interactive folder picker for me.


#6

“interactive folder picker” Sorry what does this mean?


#7


#8

ok :+1: The workflow Part…


#9

@JimmyKater When Full Text Search becomes stable (should be out of beta state pretty soon), Nextant isn’t needed anymore. The result should be pretty similar though… it will index the OCR’ed PDFs just fine; just using ElasticSearch instead of Solr, should make no difference for the end user.
Yes, I tested Full Text Search a while ago.


#10

@tv_cologne What do you mean with the “workflow part” ?


#11

@mathiasconradt

purrrrrect!
and so it makes the whole story round… yay :slight_smile:

thanks again for your great howto!


#12

My winner is about to get a PM. Thanks for playing!


#13

I used the owncloud connector and then used the syntax of: > https://<mydomain>/remote.php/dav/files/<username>/Scanbot/ and it worked.

I had tried it before with just > https://<mydomain>/ but never thought to specify the full path.

Really handy, thanks! I use Scanbot every day and was emailing docs manually…horrible!


#14

Be sure to take a moment and request Scanbot add support for multiple Nextcloud/Owncloud/WebDAV accounts / locations. You can submit your request here via their website. All requests are considered based on popularity: Make your voice heard!