Detectng faint inscriptions on rock faces

Hi all.

I came across a problem of mapping faint inscriptions on a rock
face, more precisely a weathered head stone on a church yard.

The stone has been left unattended for, say, 50 years and the
painted inscription has eroded away. The relief carved text has
been covered for some time (decades) by lichen and moss;
for all I know, it might even have been covered by earth
at some time.

During resturation works, the question has come up about
restoring the inscription. The problem is that no one knows
exactly what was written there in the first place, and the
text has eroded too far to be easily readable.

So the qustion is if it can be read by "complicated" (as
opposed to "easy") methods.

What, if any, are the options for image processing methods?
Available equipments includes standard-issue digital cameras,
torches and strobe flashes.

All and any suggestions are highly appreciated.

Rune

Rune Allnor wrote:

> I came across a problem of mapping faint inscriptions on a rock
> face, more precisely a weathered head stone on a church yard.
...
> What, if any, are the options for image processing methods?
> Available equipments includes standard-issue digital cameras,
> torches and strobe flashes.

Need to see what we are up against.  Can you
stuff an image on a website somewhere, and
post the link here?

To highlight the remaining relief, use flash
at a shallow angle.  To highlight any painted
inscription, use flash at high angle but
avoid glare.

-- 
Regards,
Martin Leese
E-mail: please@see.Web.for.e-mail.INVALID
Web: http://members.tripod.com/martin_leese/

On 20 Apr, 19:45, Martin Leese <ple...@see.Web.for.e-mail.INVALID>
wrote:
> Rune Allnor wrote:
> > I came across a problem of mapping faint inscriptions on a rock
> > face, more precisely a weathered head stone on a church yard.
> ...
> > What, if any, are the options for image processing methods?
> > Available equipments includes standard-issue digital cameras,
> > torches and strobe flashes.
>
> Need to see what we are up against.  Can you
> stuff an image on a website somewhere, and
> post the link here?

Unfortunately, no. I came across this problem in a city some distance
away from home. I did not start thinking about how to deal with this
until after I had left town.

> To highlight the remaining relief, use flash
> at a shallow angle.  To highlight any painted
> inscription, use flash at high angle but
> avoid glare.

Let's say one can take several pictures in relief  where the angle of
the
flash varies in several angles (illominating from top, left,
bottom...).
Does it make sense to try and process several images simultaneously?
I mean, using several images as input to produce one output image.

Rune

On Apr 20, 12:58 pm, Rune Allnor <all...@tele.ntnu.no> wrote:
> On 20 Apr, 19:45, Martin Leese <ple...@see.Web.for.e-mail.INVALID>
> wrote:
>
> > Rune Allnor wrote:
> > > I came across a problem of mapping faint inscriptions on a rock
> > > face, more precisely a weathered head stone on a church yard.
> > ...
> > > What, if any, are the options for image processing methods?
> > > Available equipments includes standard-issue digital cameras,
> > > torches and strobe flashes.
>
> > Need to see what we are up against.  Can you
> > stuff an image on a website somewhere, and
> > post the link here?
>
> Unfortunately, no. I came across this problem in a city some distance
> away from home. I did not start thinking about how to deal with this
> until after I had left town.
>
> > To highlight the remaining relief, use flash
> > at a shallow angle.  To highlight any painted
> > inscription, use flash at high angle but
> > avoid glare.
>
> Let's say one can take several pictures in relief  where the angle of
> the
> flash varies in several angles (illominating from top, left,
> bottom...).
> Does it make sense to try and process several images simultaneously?
> I mean, using several images as input to produce one output image.
>
> Rune

Of course it would be unjust to comment anything without looking at a
sample image of the inscriptions. But if you say it is difficult to
analyze visually even with human eyes, I would have little faith in
any vision algorithm to recreate or 'synthesize' missing information.

Good luck, and kudos for thinking of a unique way of employing human
and computer image understaing!!!

P.
[http://groups.google.com/group/medicalimagingscience?hl=en]

Rune:
Been there, done that.  Well not me, but someone very talented has,
and that someone is Dr. Tom Malzbender of HPLabs.  The technique works
wonderfully.  Here's your answer:
http://www.hpl.hp.com/research/ptm/index.html

Also of interest:
http://www.hpl.hp.com/news/2000/oct-dec/3dimaging.html (look at the
Sumerian tablets and try out the online java-based demo)  The real
application looks a lot better than the online demo which had to
sacrifice some things to get it to work quickly over the web.

He also used it to look at The Antikythera Mechanism, which will be
the subject of a Discovery Channel or History Channel documentary.
http://www.hp.com/hpinfo/newsroom/press/2006/061130c.html

Basically his PTM Viewer is a multi angle illumination technique like
you mentioned in your other post.  There are about 50 flash bulbs
placed all around a dome with a camera on top.  The flashes go off in
sequence and the camera captures all the individual photos.  The
photos are then reconstructed with a dizzying array of settable
parameters to acheive just the look you want.  It definitely brings
out faint inscriptions very well.  There's a picture on one of the
pages of one of the incarnations of his illumination system (imaging
dome).  There can be different versions of this depending on the size
of the object to be imaged, for example, a dome for small objects, or
a swing arm type of contraption that can view large paintings or other
objects hanging on walls.
Regards,
ImageAnalyst
P.S. The connection of your name to your interest in runes and
inscriptions is interesting.

On Apr 20, 1:23 pm, Rune Allnor <all...@tele.ntnu.no> wrote:
> Hi all.
>
> I came across a problem of mapping faint inscriptions on a rock
> face, more precisely a weathered head stone on a church yard.
>
> The stone has been left unattended for, say, 50 years and the
> painted inscription has eroded away. The relief carved text has
> been covered for some time (decades) by lichen and moss;
> for all I know, it might even have been covered by earth
> at some time.
>
> During resturation works, the question has come up about
> restoring the inscription. The problem is that no one knows
> exactly what was written there in the first place, and the
> text has eroded too far to be easily readable.
>
> So the qustion is if it can be read by "complicated" (as
> opposed to "easy") methods.
>
> What, if any, are the options for image processing methods?
> Available equipments includes standard-issue digital cameras,
> torches and strobe flashes.
>
> All and any suggestions are highly appreciated.
>
> Rune

On Apr 20, 3:51 pm, ImageAnalyst <imageanal...@mailinator.com> wrote:
> Rune:
> Been there, done that.  Well not me, but someone very talented has,
> and that someone is Dr. Tom Malzbender of HPLabs.  The technique works
> wonderfully.  Here's your answer:http://www.hpl.hp.com/research/ptm/index.html
>
> Also of interest:http://www.hpl.hp.com/news/2000/oct-dec/3dimaging.html(look at the
> Sumerian tablets and try out the online java-based demo)  The real
> application looks a lot better than the online demo which had to
> sacrifice some things to get it to work quickly over the web.
>
> He also used it to look at The Antikythera Mechanism, which will be
> the subject of a Discovery Channel or History Channel documentary.http://www.hp.com/hpinfo/newsroom/press/2006/061130c.html
>
> Basically his PTM Viewer is a multi angle illumination technique like
> you mentioned in your other post.  There are about 50 flash bulbs
> placed all around a dome with a camera on top.  The flashes go off in
> sequence and the camera captures all the individual photos.  The
> photos are then reconstructed with a dizzying array of settable
> parameters to acheive just the look you want.  It definitely brings
> out faint inscriptions very well.  There's a picture on one of the
> pages of one of the incarnations of his illumination system (imaging
> dome).  There can be different versions of this depending on the size
> of the object to be imaged, for example, a dome for small objects, or
> a swing arm type of contraption that can view large paintings or other
> objects hanging on walls.
> Regards,
> ImageAnalyst
> P.S. The connection of your name to your interest in runes and
> inscriptions is interesting.
>
> On Apr 20, 1:23 pm, Rune Allnor <all...@tele.ntnu.no> wrote:
>
>
>
> > Hi all.
>
> > I came across a problem of mapping faint inscriptions on a rock
> > face, more precisely a weathered head stone on a church yard.
>
> > The stone has been left unattended for, say, 50 years and the
> > painted inscription has eroded away. The relief carved text has
> > been covered for some time (decades) by lichen and moss;
> > for all I know, it might even have been covered by earth
> > at some time.
>
> > During resturation works, the question has come up about
> > restoring the inscription. The problem is that no one knows
> > exactly what was written there in the first place, and the
> > text has eroded too far to be easily readable.
>
> > So the qustion is if it can be read by "complicated" (as
> > opposed to "easy") methods.
>
> > What, if any, are the options for image processing methods?
> > Available equipments includes standard-issue digital cameras,
> > torches and strobe flashes.
>
> > All and any suggestions are highly appreciated.
>
> > Rune- Hide quoted text -
>
> - Show quoted text -

Good research, ImageAnalyst!

Well thanks for the compliment!  But it was not research for his
question.  I was just something I already knew fortuitiously and it
just happened to come in handy today.  By the way, Tom is a nice guy
and he is interested in furthering use of this technology and may (or
may not -- I won't speak for him) be interested in helping some
research groups get it up and going themselves.  I know some hobbyists
have even home-built some systems like his.
Regards,
ImageAnalyst

=====================================================
On Apr 20, 7:27 pm, "pixel.to.life" <pixel.to.l...@gmail.com> wrote:
> On Apr 20, 3:51 pm, ImageAnalyst <imageanal...@mailinator.com> wrote:
>
>
> > Rune:
> > Been there, done that.  Well not me, but someone very talented has,
> > and that someone is Dr. Tom Malzbender of HPLabs.  The technique works
> > wonderfully.  Here's your answer:http://www.hpl.hp.com/research/ptm/index.html
>
> > Also of interest:http://www.hpl.hp.com/news/2000/oct-dec/3dimaging.html(lookat the
> > Sumerian tablets and try out the online java-based demo)  The real
> > application looks a lot better than the online demo which had to
> > sacrifice some things to get it to work quickly over the web.
>
> > He also used it to look at The Antikythera Mechanism, which will be
> > the subject of a Discovery Channel or History Channel documentary.http://www.hp.com/hpinfo/newsroom/press/2006/061130c.html
>
> > Basically his PTM Viewer is a multi angle illumination technique like
> > you mentioned in your other post.  There are about 50 flash bulbs
> > placed all around a dome with a camera on top.  The flashes go off in
> > sequence and the camera captures all the individual photos.  The
> > photos are then reconstructed with a dizzying array of settable
> > parameters to acheive just the look you want.  It definitely brings
> > out faint inscriptions very well.  There's a picture on one of the
> > pages of one of the incarnations of his illumination system (imaging
> > dome).  There can be different versions of this depending on the size
> > of the object to be imaged, for example, a dome for small objects, or
> > a swing arm type of contraption that can view large paintings or other
> > objects hanging on walls.
> > Regards,
> > ImageAnalyst
> > P.S. The connection of your name to your interest in runes and
> > inscriptions is interesting.
>
> > On Apr 20, 1:23 pm, Rune Allnor <all...@tele.ntnu.no> wrote:
>
> > > Hi all.
>
> > > I came across a problem of mapping faint inscriptions on a rock
> > > face, more precisely a weathered head stone on a church yard.
>
> > > The stone has been left unattended for, say, 50 years and the
> > > painted inscription has eroded away. The relief carved text has
> > > been covered for some time (decades) by lichen and moss;
> > > for all I know, it might even have been covered by earth
> > > at some time.
>
> > > During resturation works, the question has come up about
> > > restoring the inscription. The problem is that no one knows
> > > exactly what was written there in the first place, and the
> > > text has eroded too far to be easily readable.
>
> > > So the qustion is if it can be read by "complicated" (as
> > > opposed to "easy") methods.
>
> > > What, if any, are the options for image processing methods?
> > > Available equipments includes standard-issue digital cameras,
> > > torches and strobe flashes.
>
> > > All and any suggestions are highly appreciated.
>
> > > Rune- Hide quoted text -
>
> > - Show quoted text -
>
> Good research, ImageAnalyst!- Hide quoted text -
>
> - Show quoted text -

> > > > What, if any, are the options for image processing methods?
> > > > Available equipments includes standard-issue digital cameras,
> > > > torches and strobe flashes.
>
Consider also alternate light wavelengths i.e ultraviolet and
infrared..
Mark

How would that work?  If it's rock on top (the flat, cut surface), and
the same rock in the carved/etched indentations, why would there be
any difference in percent reflectance?  It's not like there's ink in
there or some different material.  Only way I see is if there may
possibly some kind of biological material (lichen, moss) in the
indents (that's not on the main surface) but since the indentations
are so faint and shallow I doubt this is the case.  But I don't know
-- maybe you have some reasonable theory for why this might work.
Open to new ideas,
ImageAnalyst

On Apr 20, 9:09 pm, Mark <makol...@yahoo.com> wrote:
> > > > > What, if any, are the options for image processing methods?
> > > > > Available equipments includes standard-issue digital cameras,
> > > > > torches and strobe flashes.
>
> Consider also alternate light wavelengths i.e ultraviolet and
> infrared..
> Mark

On 21 Apr, 00:51, ImageAnalyst <imageanal...@mailinator.com> wrote:
> Rune:
> Been there, done that.  
...
> Basically his PTM Viewer is a multi angle illumination technique like
> you mentioned in your other post.  There are about 50 flash bulbs
> placed all around a dome with a camera on top.  The flashes go off in
> sequence and the camera captures all the individual photos.  The
> photos are then reconstructed with a dizzying array of settable
> parameters to acheive just the look you want.  It definitely brings
> out faint inscriptions very well.  

Thanks, this was the sort of thing I had in mind. I suppose this can
be one with amateur equipment? A camera on a tripod, remote flash to
be positioned for each shot... elaborate, but it ought to be possible.

> P.S. The connection of your name to your interest in runes and
> inscriptions is interesting.

...but purely coincidential. The several versions of the etymology
behind the name agree that it points to the norse rune alphabet,
but that's about all. Early etymologic intrepretations said the name
meant "he who has [hidden] insight" or "he who knows secrets",
the interpretation alluding to that early rune inscriptions (on large
bautas, 5th to 12th centuries) was clearly an instrument of the
social elite. More recent interpretations point more towards the
name meaning "he who is literate" basing the interetation on
younger finds (~14th to 15th centuries) of wooden sticks with
rune inscriptions of more mundane nature than the bautas; no
old inscriptions on wood has been preserved for the simple fact
that wooden sticks tend to rot away.

Whatever interpretation of te etymology one chooses to believe
in, it is highly debatable whether any of the above fit on me...

Rune