Import photographed documents (rotate/crop/color corrections/metadata)
I, as many others, do not have a scanner at hand anymore. I would therefore like to enable paperworks to handle photographed documents. For my use cases (e.g. keeping a receipt in case of warranty, keeping track of expenses), the typically low quality is sufficient. However, OCR and labeling is a must.*
These photos typically need some processing: Rotating for the OCR to work as a minimum; Perspective correction and color/contrast improvement would be nice.
As far as I could see from a quick evaluation, Paperworks does not do this as of now but it should be fairly straightforward to implement. To this account I would add another option besides [Scan, Scan from feeder, Import] which would be "Import photo" (or similar) and a corresponding command "import_photo" (or a switch for "import") for the cli.
If you're not generally opposed to this proposal, I would try to get a working prototype finished in the next two weeks. Please let me know if you approve
Finally: I feel comfortable with Python and the Paperworks code looks pretty well structured. I don't think I'll have any issues finding the right place to implement this. I'm of course still happy for any hints.
Cheers, Hannes
`* As a matter of fact, I'm trying to automate as much of the process to the point its get the invoice -> photograph it -> throw it away. But this is another topic that I will share with you, given I succeed
PS: Of course I will first research if this is trivial to implement because e.g. tesseract already supports its. Please also share if you have useful hints