|
Procedure for improving legibility of old records?
Sat, 22 Oct 2005 20:49:39 +1000
soc.genealogy.computing
previous
tim sewell...
|
Do any listers know of good references which offer guidance on improving the
legibility of old records?
Other than the generalised comment that "legibility can be improved by most
photoediting software", I have been unable to find any specific "how-to"
type advice.
I have tried Photoshop, Paintshop pro etc, but I am really working on the
basis of pure guesswork and trial-and-error. Basically, I "fiddle about"
with brightness, contrast, threshold setting, curves etc, without much of a
clue. And I know there must be a better way of doing things than that!
Robert Heiling...
|
I don't have the answer that you're looking for as that's what I do also
including negative image. What works varies according to the image. It's
worth emphasizing that these programs can't bring out detail that simply
isn't there to begin with.
|
I am fortunate enough to have access to a FHC which has a film scanner
installed, so that we can copy direct to CD-Rom. But even then, there are
areas of varying intensity where I can just read a name on the film, but not
on the scanned record. The sort of problem that puzzles me most is where
Robert Heiling...
|
Something is terribly wrong with the scan in that case. You could take a
look at the image information for those images on CD (Paintshop, click
View-Image Information). For display purposes at original size, those
may be set at 72dpi which doesn't capture enough detail.
|
varying fading of the original ink has caused faint portions within a
record. eg one of my ancestors was Ann Tait. The "A" is clearly legible,
the "nn" that follows it can only just be read.
Robert Heiling...
|
I find that scanning a document is a more effective way to make out
characters on a document than a magnifying glass, but the resolution
needs to be there. I might scan something like that at 600dpi or better
to check results. If the scan is not as legible as the image in the
viewer and there is no resolution option when going to CD, then I would
opt to print the document and scan it yourself at a better density.
|
None of the reference books that I have seen so far are much help at all,
being aimed at photographers, and mostly colour photographers at that. If
no such references are accessible, what do other listers do for their own
records?
I would welcome any/all advice from other listers.
Robert Heiling...
|
Just remember that if it isn't there to begin with, no magic can bring
it out. :-)
|
mickg...
|
Using Photoshop try Autolevels then autocontrast. Then using selection
tools Pick out areas that need improvement and use levels, brightess and
contrast carefully on just the area, Area selection which can be a block
or a custome selected shape makes the levels etc settings happen only in
that area.
ALWAYS do this on a copy keeping the original as it was as a separate file.
|
singhals...
|
As others have said, if it's not on the film itself, it won't be on the
scan. And, as they have also said, trial and error is the ONLY way
you'll get good results for the section of the page that you really
really want to read.
Re the film scan at the FHC -- the viewer has a light control knob on
the front -- use it without regard to the markings which are color-coded
to the lens. Some of those settings put a darker spot in the middle of
the page to compensate for the bull's=eye lighting on the film, some
light the entire page evenly, some put a brighter spot in the middle of
the page.
ALWAYS use the grey-scale imaging, and in some applications on the
thing, the JPG is larger than the TIFF because the JPG is color and the
TIFF is a compressed B&W. This is important because often color scans
can be more satisfactorily tweaked than B&W/grey-scales.
To be more helpful than that about that, I'd need to know the brand
names of the hardware involved *and* the exact name of the program
you're using to make your scans.
|
|
next
|