Daniel Reetz, the founder of the DIY Book Scanner community, has recently started making videos of prototyping and shop tips. If you are tinkering with a book scanner (or any other project) in your home shop, these tips will come in handy. https://www.youtube.com/channel/UCn0gq8 ... g_8K1nfInQ

Scan Tailor "Experimental"

Scan Tailor specific announcements, releases, workflows, tips, etc. NO FEATURE REQUESTS IN THIS FORUM, please.
Tulon
Posts: 687
Joined: 03 Oct 2009, 06:13
Number of books owned: 0
Location: London, UK
Contact:

Re: Scan Tailor "Experimental"

Post by Tulon » 23 Aug 2015, 08:47

VipeRsan wrote:
Tulon wrote:Can't do. Because ST no longer asks users to correct DPIs, it can no longer map between physical units and pixels.
Then maybe in pixels? It should better to align pages with constant amount of pixels than calculate based on percentage.
Pixels won't work well with "Match size by scaling" mode.
VipeRsan wrote: Also it would be good to use canvas size adapted to the standard ratio of A4, A5, B4, B5, letter, custom, etc. It can be done based on simple proportion to calculate amount of pixels in x and y direction.
ST is primarily targeted at making eBooks by non-professionals. Such a feature would not be terribly useful to them. Besides, these days I only work on ST for fun, which means I don't really accept feature requests.
Scan Tailor experimental doesn't output 96 DPI images. It's just what your software shows when DPI information is missing. Usually what you get is input DPI times the resolution enhancement factor.

d14b0ll0s
Posts: 31
Joined: 17 Aug 2015, 19:37
Number of books owned: 3000
Country: Poland

Re: Scan Tailor "Experimental"

Post by d14b0ll0s » 23 Aug 2015, 10:02

I was able to compile it under OS X, but when I try to add any files to the project, the second I click [OK] the names turn red and I get the error message in a popup window saying:
Some of the files failed to load.
Either we don't support their format, or they are broken.
You should remove them from the project.
I was trying to add the same JPEG files that I used before in the STE running in a VM. I also tried with different folders, with the same effect.
Does scantailor log somewhere by default? I see that there is a 'Debug mode' in the Tools menu. Sorry if this is obvious or documented, I did not have a chance to dig into this yet.

Tulon
Posts: 687
Joined: 03 Oct 2009, 06:13
Number of books owned: 0
Location: London, UK
Contact:

Re: Scan Tailor "Experimental"

Post by Tulon » 23 Aug 2015, 10:23

I was trying to add the same JPEG files that I used before in the STE running in a VM. I also tried with different folders, with the same effect.
Does scantailor log somewhere by default? I see that there is a 'Debug mode' in the Tools menu. Sorry if this is obvious or documented, I did not have a chance to dig into this yet.
It doesn't log anywhere but the console, and even there it logs very little.
The code you want to debug is in JpegMetadataLoader.cpp. Unfortunately I didn't bother implementing various error message callbacks to be able to capture error messages coming from the underlying jpeg library.
Scan Tailor experimental doesn't output 96 DPI images. It's just what your software shows when DPI information is missing. Usually what you get is input DPI times the resolution enhancement factor.

d14b0ll0s
Posts: 31
Joined: 17 Aug 2015, 19:37
Number of books owned: 3000
Country: Poland

Re: Scan Tailor "Experimental"

Post by d14b0ll0s » 23 Aug 2015, 10:27

Tulon wrote:It doesn't log anywhere but the console, and even there it logs very little.
The code you want to debug is in JpegMetadataLoader.cpp. Unfortunately I didn't bother implementing various error message callbacks to be able to capture error messages coming from the underlying jpeg library.
So it seems we have little chance of knowing what's going on?
I'll try to play with the commandline version in a couple of days and see what happens there.

Tulon
Posts: 687
Joined: 03 Oct 2009, 06:13
Number of books owned: 0
Location: London, UK
Contact:

Re: Scan Tailor "Experimental"

Post by Tulon » 23 Aug 2015, 11:13

The command-line version in STE is completely broken, so that won't get you anywhere. At this point an actual developer running OSX is necessary to debug this issue.
Scan Tailor experimental doesn't output 96 DPI images. It's just what your software shows when DPI information is missing. Usually what you get is input DPI times the resolution enhancement factor.

d14b0ll0s
Posts: 31
Joined: 17 Aug 2015, 19:37
Number of books owned: 3000
Country: Poland

Re: Scan Tailor "Experimental"

Post by d14b0ll0s » 23 Aug 2015, 11:33

Tulon wrote:The command-line version in STE is completely broken, so that won't get you anywhere. At this point an actual developer running OSX is necessary to debug this issue.
Sorry I can't be of more help due to the lack of skills. I'll be happy to try what I can though if you need to check something.

Adam32
Posts: 28
Joined: 28 Jun 2014, 08:55
Number of books owned: 500
Country: United Kingdom

Re: Scan Tailor "Experimental"

Post by Adam32 » 28 Aug 2015, 04:36

This is absolutely fantastic, I just can't believe the speed increases. Is there any chance of adding this to Linux repositories?

Tulon
Posts: 687
Joined: 03 Oct 2009, 06:13
Number of books owned: 0
Location: London, UK
Contact:

Re: Scan Tailor "Experimental"

Post by Tulon » 28 Aug 2015, 04:57

Adam32 wrote:This is absolutely fantastic, I just can't believe the speed increases. Is there any chance of adding this to Linux repositories?
Before it can be added to Linux repositories, at the very minimum the command-line version will need to be fixed. That's not exactly easy or fun, so don't expect this too happen soon.
Scan Tailor experimental doesn't output 96 DPI images. It's just what your software shows when DPI information is missing. Usually what you get is input DPI times the resolution enhancement factor.

abmartin
Posts: 79
Joined: 15 Sep 2010, 15:33
Number of books owned: 2000
Country: USA
Location: Ohio

Re: Scan Tailor "Experimental"

Post by abmartin » 04 Sep 2015, 21:25

Thanks for the return with some nice speed updates! I'm loving the new post-DPI Scantailor paradigm. It's easy enough to add the DPI info in afterwards if I want to. (For anyone wondering, just use imagemagick: mogrify -density ### -units PixelsPerInch *.tif) I didn't think I could be more impressed by Tulon, but it turns out I was wrong and am even more impressed! I haven't had a chance to really play with the new geometry, but I like the greater cleanliness of the approach. I am just astonished. The speed bump is very welcome.

A quick note to VipeRsan about getting things to a target output size. I've hacked together a little linux bash script that will do that over on an older tutorial thread. Since all the tools are cross-platform, I'm sure that the same approach would work in the OS of your choice. http://diybookscanner.org/forum/viewtop ... =19&t=2747

Tulon
Posts: 687
Joined: 03 Oct 2009, 06:13
Number of books owned: 0
Location: London, UK
Contact:

Re: Scan Tailor "Experimental"

Post by Tulon » 07 Sep 2015, 18:31

A new release is ready: https://github.com/Tulon/scantailor/rel ... 2015_09_07
This version should be even faster on the Output stage when using OpenCL.
Scan Tailor experimental doesn't output 96 DPI images. It's just what your software shows when DPI information is missing. Usually what you get is input DPI times the resolution enhancement factor.

Post Reply