Jump to content

Scanning and/or using Gimp to improve photos?



gawain3

Recommended Posts

Hope I'm in the correct forum, if not would a moderator move my question to the correct forum.

 

We've all noticed that photographs of ink samples on paper often turn out grey or blue. For example...

 

Image.thumb.jpeg.cf2fe968339c00c03d196fa98e0ba318.jpeg

 

Samples scanned are of course much better examples of the color of the reviewed ink.

 

1210127224_Image3.thumb.jpeg.1247a193523fd2d90a06134a4312438a.jpeg

Not only better, but a scanner removes the show through from the other side.

 

Another example of improved color is quickly running the photo through Gimp, a Linux photo editor which also has a Windows version.

1442191325_Image2.thumb.jpeg.3adca5f276e6b828ce6a136154a464c4.jpeg

 

 

So I wonder why people don't scan or use Gimp to improve their photos. If you don't have a scanner, Gimp is a free download that will visibly improve your photos and give your readers a much better idea of the colors in your photos. Now this app might seem complicated - and, yes, it does have a large learning curve - but all you do is go to Colors/Levels find the highlights tool and whiten the paper and lighten the overly dark ink. Here is a screenshot of Gimp.

Screenshot.thumb.png.01dfa3424a875383982052d2a1616b65.png

 

 

I not trying to insult anyone or be a true PITA, but doesn't everyone want more accurate pictures in your post?

 

P.S. These are very quick scans and use of Gimp. A little practice and you can do better.

 

Image 2.jpeg

Screenshot.png

Link to post
Share on other sites
  • Replies 26
  • Created
  • Last Reply

Top Posters In This Topic

  • gawain3

    7

  • lapis

    3

  • A Smug Dill

    3

  • LizEF

    3

Top Posters In This Topic

Posted Images

Ink Stained Wretch

I've been using GIMP for photographs and more for quite a number of years. I tend to treat everything I post on-line, here and elsewhere as grist for GIMP.

 

I don't know that most people on-line these days are into more than the very convenient tradition of take a snapshot, post it, see what the reaction is.

 

Sure, let people work on images before they post them. Some of us will and some of us won't.

 

BTW, for those who want to see what we're talking about here's the GIMP Web site.

On a sacred quest for the perfect blue ink mixture!

ink stained wretch filling inkwell

Link to post
Share on other sites

Thanks for this post and yes, this is the right sub-forum.

But scans are not always better than photos. Scans themselves and then their reproductions (including Microsoft's old-age Office Picture Manager and Paint) can be misleading. Your examples above are IMO also misleading. In your first scan (up top here), the Pilot Blue looks less blue, less saturated and more green than in the photo. How can we tell which is "better"? Also, the avoidance of show-though is although of course more encouraging, less realistic. If show-through is there, I want to see it and not want to have it hidden. Gimp (which I don't have) may be a bit better (especially after practicing with it) but even that can still look different from the real ink on real paper. Our computer screens are usually different, too. Duh....

 

Life is too short to drink bad wine (Goethe)

Link to post
Share on other sites
2 hours ago, gawain3 said:

So I wonder why people don't scan or use Gimp to improve their photos.


How can you tell ?

 

i don’t know what Gimp is but I modify my photos with the standard photo editor on my iPad. I adjust the color of the ink on a photo to make it look the same as the ink on the original. I usually don’t bother to adjust the color of the paper.

Link to post
Share on other sites
8 hours ago, gawain3 said:

So I wonder why people don't scan or use Gimp to improve their photos.

 

Maybe they don't feel, or think, or find it — intuitively, or by more rigorous cost-benefit analysis, or judging from experience — worthwhile to take the time and make the effort to do so. How does improving their photos, specifically in the way you suggested, make the act of posting and sharing more rewarding to them?

 

8 hours ago, gawain3 said:

I not trying to insult anyone or be a true PITA, but doesn't everyone want more accurate pictures in your post?

 

I can't speak for everyone or even anyone else, but no, I don't always want more ‘accurate’ pictures in my post, when each image or figure is attached or embedded in the post to illustrate one or more points I'm making. I don't doctor photos in order to deliberately misrepresent some aspect of the subject matter, and I don't set out to deceive others; but I also don't feel compelled to present information (including pictorially) in the way particular readers prefer it, address their specific interests and/or concerns, or generally to make it easier for them to reconcile what they see with their individual cognitive frameworks.

 

By the way, almost all of the time I do use GIMP to edit images — whether they were shot on digital camera or scanned from a page — before posting them, but it has nothing to do with wanting the images to be more ‘accurate’ by anybody else's standard.

 

 

Oh, and doesn't everyone want their discussion thread title to be more informative about its subject matter?

I endeavour to be frank and truthful in what I write, show or otherwise present, when I relate my first-hand experiences that are not independently verifiable; and link to third-party content where I can, when I make a claim or refute a statement of fact in a thread. If there is something you can verify for yourself, I entreat you to do so, and judge for yourself what is right, correct for valid. I may be wrong, and my position or say-so is no more authoritative and carries no more weight than anyone else's here.

Link to post
Share on other sites

The main problem is that the original image is badly underexposed.

 

Camera light meters are calibrated to read correctly for an "average" scene. Images which are predominantly light colours (such as a sheet of white paper with a writing sample) which reflect much more light than the camera is calibrated to expect will end up underexposed unless adjustments are made. If your camera has a custom mode for snow scenes, using that will give you a much better starting point, as would dialing at least one stop of positive exposure compensation. The lighting under which the picture is taken may also introduce a colour cast to the image (white balance). All the post-processing adjustments discussed are just trying to compensate for these errors. 

 

See here for a rundown of how cameras can get exposure wrong -  https://tinyurl.com/ef9hadfy , and here for issues relating to white balance - https://www.techradar.com/news/the-a-to-z-of-photography-white-balance.

=====================================
Mario Mirabile
Melbourne, Australia

www.miralightimaging.com

=====================================
Link to post
Share on other sites
  • lapis changed the title to Scanning and/or using Gimp to improve photos?
9 hours ago, A Smug Dill said:

Oh, and doesn't everyone want their discussion thread title to be more informative about its subject matter?

Fixed.

Life is too short to drink bad wine (Goethe)

Link to post
Share on other sites
20 hours ago, Ink Stained Wretch said:

I've been using GIMP for photographs and more for quite a number of years. I tend to treat everything I post on-line, here and elsewhere as grist for GIMP.

 

I don't know that most people on-line these days are into more than the very convenient tradition of take a snapshot, post it, see what the reaction is.

 

Sure, let people work on images before they post them. Some of us will and some of us won't.

 

BTW, for those who want to see what we're talking about here's the GIMP Web site.

Thank you for that, I should have included the link.

Link to post
Share on other sites
20 hours ago, lapis said:

Thanks for this post and yes, this is the right sub-forum.

But scans are not always better than photos. Scans themselves and then their reproductions (including Microsoft's old-age Office Picture Manager and Paint) can be misleading. Your examples above are IMO also misleading. In your first scan (up top here), the Pilot Blue looks less blue, less saturated and more green than in the photo. How can we tell which is "better"? Also, the avoidance of show-though is although of course more encouraging, less realistic. If show-through is there, I want to see it and not want to have it hidden. Gimp (which I don't have) may be a bit better (especially after practicing with it) but even that can still look different from the real ink on real paper. Our computer screens are usually different, too. Duh....

 

 

But wouldn't you agree the a scan or a gimp treated photo often gives the reader a much better idea of the ink's true color?

 

As for the show through, my only paper that has both sides clean is still in the mail.

Link to post
Share on other sites
17 hours ago, mariom said:

The main problem is that the original image is badly underexposed.

 

Camera light meters are calibrated to read correctly for an "average" scene. Images which are predominantly light colours (such as a sheet of white paper with a writing sample) which reflect much more light than the camera is calibrated to expect will end up underexposed unless adjustments are made. If your camera has a custom mode for snow scenes, using that will give you a much better starting point, as would dialing at least one stop of positive exposure compensation. The lighting under which the picture is taken may also introduce a colour cast to the image (white balance). All the post-processing adjustments discussed are just trying to compensate for these errors. 

 

See here for a rundown of how cameras can get exposure wrong -  https://tinyurl.com/ef9hadfy , and here for issues relating to white balance - https://www.techradar.com/news/the-a-to-z-of-photography-white-balance.

 

Again, the picture was taken with an iPhone. I used to take all my photos with an Olympus 4. Been in a closet for a while, I'm afraid.

Link to post
Share on other sites
22 minutes ago, gawain3 said:

But wouldn't you agree the a scan or a gimp treated photo often gives the reader a much better idea of the ink's true color?

 

As for the show through, my only paper that has both sides clean is still in the mail.

Diamine Salamander.  All images made by me, as they were captured, no attempt at doctoring. Salamander is a murky green (see linked name).  It is not what the scanner images show - nowhere near what they show.  Please note that other colors - like Waterman Audacious Red and Inspired Blue - scan accurately.  Note too that I've just recently scanned 93 of these pages.  Scanners are not miraculous color reproducers.

 

1. Screen capture from a video

large.DiamineSalamander1.jpg.b5cacf8726ce8d619f420a41fdaab402.jpg

 

2. Photograph with the same phone as made the video (+2 on the camera app's exposure setting, though it's not entirely clear whether hitting the + button increase exposure by .5 or 1)

large.DiamineSalamander2.jpg.8eacf3f66d1da6e4bb020f000b2023e8.jpg

 

3. Scan with a black page (construction paper) covering the sheet, no attempt to use the scanning software to change defaults

large.DiamineSalamander3a.jpg.4ef38603f7c6863b8062311bc783d4cd.jpg

 

3b. Scan as 3, but choosing the scanner's "Restore faded color" option

large.DiamineSalamander3b.jpg.ffe83e04145f7e721855e36e87513a4d.jpg

 

4. Scan without the black page (just the lid of the scanner - you can see the negative strip reader in the background starting behind the "R" in Rhodia on the first line)

large.DiamineSalamander4a.jpg.d460e9f862e352bc5e67aa75443a9787.jpg

 

4b. As 4, but "Restore faded color"

large.DiamineSalamander4b.jpg.57d5532c126a113133c6dbf730153010.jpg

 

Your "turn the page white" instructions for GIMP will not help these images tell the user what Salamander looks like.

Link to post
Share on other sites

Colour correcting images is a tricky business. If ones monitor is not colour calibrated it's just wishing and hoping. Any ambient light other than D55-60 will skew the result, as will the colour of the walls in the room

 

Including a grey scale in the image can be helpful for both the person correcting the image, as well as those looking at said image who know what to do with it.

 

As for "turn the paper white", most papers have a tint, and are, therefore, not white, so taking this approach will change the colour of the ink on the page as well.

 

After all that, most viewing the resultant files won't have a calibrated monitor.

 

Tricky business.

 

Then there's calibrating the scanner to match the monitor.

"Simplicate and add Lightness."

Link to post
Share on other sites

Again, I’m just trying to get a better image of the ink. As for gimp, I’m recommending it as a tool to make a better image than my iPhone.

Link to post
Share on other sites
3 hours ago, gawain3 said:

Again, I’m just trying to get a better image of the ink.

 

… essentially by telling prompting contributors of ink reviews to do some more work, but couching it as, ”Wouldn't you want to as a matter of course?”

 

I gratefully learnt from fellow members, such as @XYZZY and @BaronWulfraed, who freely shared similar suggestions when interested folk such as @LizEF and myself were discussing ink review methodology (oh my, how time flies!) — because clearly we wanted to know, and even now we're still exploring.

 

If you're commenting from the position of being a consumer of images and reviews posted by others, though, I'd prefer you to couch your suggestions as requests from which you intend to be the beneficiary — even just by being a member of a homogenous class of many beneficiaries — for the doers to consider, instead of making it sound as if what they do today is inadequate for what they want for themselves, and that they see the need for self-improvement (but just didn't know how, and could use some guidance).

I endeavour to be frank and truthful in what I write, show or otherwise present, when I relate my first-hand experiences that are not independently verifiable; and link to third-party content where I can, when I make a claim or refute a statement of fact in a thread. If there is something you can verify for yourself, I entreat you to do so, and judge for yourself what is right, correct for valid. I may be wrong, and my position or say-so is no more authoritative and carries no more weight than anyone else's here.

Link to post
Share on other sites

I first used a computer over 30 years ago.  I took to them fairly easily and eventually found myself (became) a software developer building database applications.

 

My aunt, who's in her 80s, has never been able to grasp computer concepts (particularly not principles as opposed to literal, step-by-step instructions), and it drives me insane to try to teach her how to do something which to me seems so easy and obvious, but which she cannot grasp.

 

I expect it drives me crazy because I cannot remember how ignorant I was before I first used a computer.

 

When it comes to color correction, I'm my clueless aunt, and the people claiming it's easy are programmer-me who've long forgotten their past ignorance, and we're both really frustrated with each other.

Link to post
Share on other sites
5 hours ago, LizEF said:

When it comes to color correction, I'm my clueless aunt, and the people claiming it's easy are programmer-me who've long forgotten their past ignorance, and we're both really frustrated with each other.

When it comes to color correction, I'm my clueless grandfather (who I hardly met and in those days I wasn't even in school where I had to learn how to use a fountain pen with Skrip ink). But even if I was -- and even if he and I had long talked about inks -- his conception and/or interpretation about color, wetness, show-through, and permanence...... that would never have been discussable -- let alone judgeable -- over, under, around and through my computer.

 

Life is too short to drink bad wine (Goethe)

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now







×
×
  • Create New...