1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
|
#+EXPORT_FILE_NAME: README
* Scanner: scan documents and images with Emacs
Scan documents and images using =scanimage(1)= from the SANE distribution and
=tesseract(1)= for OCR and PDF export. Additionally, =unpaper(1)= can now be
used for post-processing the scans obtained from =scanimage= before feeding
them into =tesseract=. This is optional, but highly recommended. The source
to unpaper is available at https://github.com/unpaper/unpaper.
The scanner package uses two sets of customizations for image mode and
document mode, with the former usually configured to use high resolution and
an image file format, like JPEG, and the latter to use lower resolution and
a document format, like PDF or text. The available file formats are
provided by =scanimage(1)= for image mode and =tesseract(1)= for document
mode. The scanner package uses =tesseract(1)= to provide optical character
recognition (OCR). You can select the language plugins with
=scanner-tesseract-languages=. See also the remark about the data
directories below.
In document mode, you can scan one or multiple pages that are then written
in a customizable output format, e.g. (searchable) PDF or text, or whatever
tesseract provides. You can also customize resolution, intermediate image
format, and paper size. The command =scanner-scan-document= starts a
document scan. Without a prefix argument, it scans one page. With a
non-numeric argument, it asks the user after each scanned page for
confirmation to scan another page. With a numeric argument, it scans that
many pages. In the latter case, it observes a delay between scans that is
customizable using =scanner-scan-delay=.
The =scanner-scan-image= command performs one scan or multiple scans in
image mode. This function tries to guess the file format from the chosen
file name or falls back to the configured default, see
=scanner-image-format=. The prefix argument works as in document mode.
The scanning commands are also available in the Tools->Scanner menu.
For both images and documents, you can customize the scan mode (e.g. "Color"
or "Gray") if your scanning device supports it.
You can pass additional options to the backends using the customization
variables =scanner-scanimage-switches= and =scanner-tesseract-switches=.
The former variable is helpful for tuning brightness and contrast, for
instance.
Finally, the customization options =scanner-tessdata-dir= and
=scanner-tessdata-configdir= must be set to point to tesseract's data
directory containing the language definitions (usually something like
=/usr/share/tessdata/=) and tesseract's configs directory containing the
output configurations (usually something like
=/usr/share/tessdata/configs/=).
* Configuration
To use =unpaper=, set the customization option =scanner-use-unpaper= to t.
Scanner defines a keymap that is best bound to some convenient key, for
example with
=(keymap-global-set "s-s" scanner-map)=
or when using =use-package= with
=:bind-keymap ("s-s" . scanner-map)=
in the use-package specification.
Most package options are customizations and can but configured in the usual
ways.
* Bugs
- This package doesn't support document feeders yet.
- This package doesn't support authentication.
- If a new document scan is started while another is still running, the log
will be messed up a bit.
|
