Scanning on curved surface?

[Bror Johansson]

Many years ago there was a simple type of scanner that was "rolled" by hand
over the object to be scanned.

Now I have a need to scan some labels on glassbottles, and came to think
about that old scanning-principle.

Are such scanners still being made?
Are there other means or tools that could facilitate such a scanning need?

/BJ

 

[SwissTony]

Many years ago there was a simple type of scanner that was "rolled" by hand
over the object to be scanned.

Now I have a need to scan some labels on glassbottles, and came to think
about that old scanning-principle.

Are such scanners still being made?
Are there other means or tools that could facilitate such a scanning need?

/BJ

can you not soak the labels and peel them off???

 

[Bror Johansson]

can you not soak the labels and peel them off???

Yes, of course I could do that. But, in this case I want to keep the bottle
intact with its labels (but without content, however). I did consider
removing them, and after scanning re-glueing them to the bottle, but I don't
want to take the risks involved in this particular case.
/BJ

 

[Wilfred]

Many years ago there was a simple type of scanner that was "rolled" by hand
over the object to be scanned.

Now I have a need to scan some labels on glassbottles, and came to think
about that old scanning-principle.

Are such scanners still being made?
Are there other means or tools that could facilitate such a scanning need?

/BJ

You may want to try and use a flatbed with large depth of field. Then
correct the distortion in Photoshop or another image editor: Filter >
Distort > Spherize. Set filter to 'horizontal only' or 'vertical only'
as applies and apply a negative amount of spherical (in fact,
cylindrical) distortion.

 

[Ian Riches]

Wilfred ([email protected]) wrote...

You may want to try and use a flatbed with large depth of field. Then
correct the distortion in Photoshop or another image editor: Filter >
Distort > Spherize. Set filter to 'horizontal only' or 'vertical only'
as applies and apply a negative amount of spherical (in fact,
cylindrical) distortion.

Good idea. A digital camera rather than a flatbed scanner may give
better results using this method. DOF will probably be better, but
resolution may not be as high.

Ian

 

[Martin Trautmann]

You may want to try and use a flatbed with large depth of field.

Is there any common test object in order to compare the DoF of a
scanner? I'd like to know in order to compare e.g. the differences of
sensors and illumination.

I wonder how much this does influence e.g. flatbed scanners with film
options, since a negative one or the other way around, or even a slide
could be 'out of focus' by more than one millimeter.

 

[Wilfred]

Is there any common test object in order to compare the DoF of a
scanner? I'd like to know in order to compare e.g. the differences of
sensors and illumination.

I wonder how much this does influence e.g. flatbed scanners with film
options, since a negative one or the other way around, or even a slide
could be 'out of focus' by more than one millimeter.

AFAIK, that is never a problem with flatbeds. They all have sufficient
DOF to eliminate curvatures in films. For film scanners it's different.

 

[Wilfred]

Wilfred ([email protected]) wrote...



Good idea. A digital camera rather than a flatbed scanner may give
better results using this method. DOF will probably be better, but
resolution may not be as high.

Hadn't thought of that - indeed worth to consider. Perhaps it's easier
to position the bottle on a scanner, though.

 

[Bror Johansson]

Wilfred ([email protected]) wrote...

Good idea. A digital camera rather than a flatbed scanner may give
better results using this method. DOF will probably be better, but
resolution may not be as high.

Ian

In fact, I've already tried the digital camera approach. There is one very
difficult obstacle to conquer: I tried some different illuminations, but
_always_ got some severe reflections. The labels I'm interested in are
partly metallic and highly reflectible.

I guess that the proposed scanning method could be - at least
somewhat -disturbed by reflections. It is also worth to consider the fact
that one of the labels cover half of the bottle's circumference. However I
thing I will try it, nevertheless.
/BJ

 

[Martin Trautmann]

In fact, I've already tried the digital camera approach. There is one very
difficult obstacle to conquer: I tried some different illuminations, but
_always_ got some severe reflections. The labels I'm interested in are
partly metallic and highly reflectible.

so you need some kind of light tent around for smooth and bright
illumination.


You won't get 360 degress by the camera approach obviously. You may have
to rescale and glue together, as it is done for panorama images.

However, you may simplify this task by placing some kind of flexible
scale around, e.g. two stripes of tape measure (is this the correct
English term) above and below.

 

[CSM1]

Is there any common test object in order to compare the DoF of a
scanner? I'd like to know in order to compare e.g. the differences of
sensors and illumination.

I wonder how much this does influence e.g. flatbed scanners with film
options, since a negative one or the other way around, or even a slide
could be 'out of focus' by more than one millimeter.

CCD sensor flatbeds have the greatest DOF. (Charge Coupled Device)
CIS sensor (the Canon Lide series) have zero DOF. The C is for Contact.

 

[Martin Trautmann]

CCD sensor flatbeds have the greatest DOF. (Charge Coupled Device)
CIS sensor (the Canon Lide series) have zero DOF. The C is for Contact.

I thought, C is for 'Canon' ;-)

In fact they claim CIS means 'Compact Image Sensor'
<http://www.canon.de/for_home/product_finder/scanners/technologien/lide/index.asp>


However, compact is googler than contact:

http://www.google.com/search?num=100&hl=en&lr=&q=CIS+"contact+image+sensor"&btnG=Search

is 14600 to 338 for

http://www.google.com/search?num=100&hl=en&lr=&q=CIS+"compact+image+sensor"&btnG=Search>

 

[CSM1]

I thought, C is for 'Canon' ;-)

In fact they claim CIS means 'Compact Image Sensor'
<http://www.canon.de/for_home/product_finder/scanners/technologien/lide/index.asp>


However, compact is googler than contact:

http://www.google.com/search?num=100&hl=en&lr=&q=CIS+"contact+image+sensor"&btnG=Search

is 14600 to 338 for

http://www.google.com/search?num=100&hl=en&lr=&q=CIS+"compact+image+sensor"&btnG=Search>

Another definition:
http://www.webopedia.com/TERM/C/contact_image_sensor.html

If Canon is using 'Compact Image Sensor' then it is the marketing
department's doing.

The CIS scanner is described as:
"A CIS scanner is more compact than a CCD scanner and can be used in smaller
products than CCD scanning technologies."

 

[Matthew Dranchak]

A notice of the availability of this device showed up in my e-mail.
Would it be of any use to you? It might scan on a curved surface.
I did not pursue this to any degree.
http://www.techdepot.com/product.asp?ProductID=2057313&affid=9841002

 

[Wayne Fulton]

Many years ago there was a simple type of scanner that was "rolled" by hand
over the object to be scanned.

Now I have a need to scan some labels on glassbottles

OK, maybe it's not a great idea, but you can roll a straight-sided round
bottle or can along on the flatbed scanner glass bed, in sync just barely
ahead of the scanning lamp as it scans. The scanner scans one row of pixels
at a time.. just keep the bottle where that row is being scanned. Then
depth of field wont be the issue, but smooth steady rolling motion will be.

With just a little practice, you can get pretty good at keeping the rolling
motion steady and in the correct place, and the result should be acceptable
for many purposes. The scanned circumference must be less than the length
of the scanner bed.

 

[MPA]

You may want to try and use a flatbed with large depth of field. Then
correct the distortion in Photoshop or another image editor: Filter >
Distort > Spherize. Set filter to 'horizontal only' or 'vertical only'
as applies and apply a negative amount of spherical (in fact,
cylindrical) distortion.

you need a linear scan panoramic camera like roundshot-very expensive.
a turntable synchronized with the film.
this could be done with a simple turning disc and digital cam and
stitching method. if filmcamera is used take fixed focus long lens.
maybe shooting every 10 degrees or 5 degrees. try it out.

 

[MPA]

Is there any common test object in order to compare the DoF of a
scanner? I'd like to know in order to compare e.g. the differences of
sensors and illumination.

I wonder how much this does influence e.g. flatbed scanners with film
options, since a negative one or the other way around, or even a slide
could be 'out of focus' by more than one millimeter.

andrew davidhazy has a wonderful precise theory on his RIT-site
regarding perfect shooting/scanning of objects with difficult shaped.

 

[MPA]

AFAIK, that is never a problem with flatbeds. They all have sufficient
DOF to eliminate curvatures in films. For film scanners it's different.

wrong! one scanner-technology has limited DOF. which one?

 

[MPA]

A notice of the availability of this device showed up in my e-mail.
Would it be of any use to you? It might scan on a curved surface.
I did not pursue this to any degree.
http://www.techdepot.com/product.asp?ProductID=2057313&affid=9841002

i still have these- b+w only handscanners(isa-card only) from
logitech(scanman)-resolution was not so high. maybe even if we could get
an adapter the software may not work anymore.

i think we could make a holder and let the bottle turn.
i still have these- b+w only handscanners(isa-card only) from
logitech(scanman)-resolution was not so high.

 

[Martin Trautmann]

Second post:
The Canon USA site says for the LiDE 35:Scanner Element

Contact Image Sensor (CIS)


http://consumer.usa.canon.com/ir/controller?act=ModelTechSpecsAct&fcategoryid=119&modelid=10241

I did not want to argue that Canon sometimes name it 'contact' as well.
However, I was suprised that they named it 'compact' in my area, which
is a less obvious hint for dof=zero.