ImageJ

3D Display

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

20 messages Options

Luciano Paulino da Silva

3D Display

Dear All,
We have some 3D surface image files that are not obtained from stacking
procedures (Atomic force microscopy images - Topography).
In this way, we have matrix files (512x512; rows x columns) for which the
corresponding number on the matrix is the height (z axis). How could we
open such files using ImageJ in order to obtain real surface plots? Are
there some plugin?
Luciano and Rebecca

David Hovis

Re: 3D Display

On Mar 12, 2008, at 9:57 AM, Luciano Paulino da Silva wrote:

> Dear All,
> We have some 3D surface image files that are not obtained from
> stacking
> procedures (Atomic force microscopy images - Topography).
> In this way, we have matrix files (512x512; rows x columns) for
> which the
> corresponding number on the matrix is the height (z axis). How could
> we
> open such files using ImageJ in order to obtain real surface plots?
> Are
> there some plugin?
> Luciano and Rebecca

File->Import->Text Image...

Should do the trick. You will end up with a 32-bit image. You can
then assign lookup tables, play with the Interactive 3D Surface Plot
plugin, etc.

--David

----------------------------------------
David Hovis
Senior Research Associate
Department of Materials Science
Case Western Reserve University
[hidden email]

Joachim Wesner

Antwort: 3D Display

In reply to this post by Luciano Paulino da Silva

Hi,

You mean you have a whole lot of files, each for one z-position? If the
files are numbered sequentially, you can use Import(/image sequence.!!

However, the resulting image stack will assume that those slices are spaced
uniformly. If the nonuniform z-position is somewhere encoded in the file
name or even
in a binary or text field of each file, you would need to write a (simple
but) special plugin AFAIK.

Mit freundlichen Grüßen / Best regards

Joachim Wesner

Luciano Paulino
da Silva
<lucianopaulinosi An
[hidden email]> [hidden email]
Gesendet von: Kopie
ImageJ Interest
Group Thema
<[hidden email]. 3D Display
GOV>

12.03.2008 14:57

Bitte antworten
an
ImageJ Interest
Group
<[hidden email].
GOV>

Dear All,
We have some 3D surface image files that are not obtained from stacking
procedures (Atomic force microscopy images - Topography).
In this way, we have matrix files (512x512; rows x columns) for which the
corresponding number on the matrix is the height (z axis). How could we
open such files using ImageJ in order to obtain real surface plots? Are
there some plugin?
Luciano and Rebecca

______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________

Luciano Paulino da Silva

Re: 3D Display

Dear Joachim Wesner,
Thank you very much for your suggestions.
We do not have stacked images but only real surface image. In this case, we
just have one image exhibiting all X, Y and Z points in order to be
rendered. In fact, the z-positions are nonuniform. We did not understand
what means the plugin AFAIK. Could you suggest something to do with this
problem?
Could we send you some of our images and the corresponding .txt file of such
data?
Thanks in advance,
Luciano and Rebecca

2008/3/12, Joachim Wesner <[hidden email]>:

>
> Hi,
>
> You mean you have a whole lot of files, each for one z-position? If the
> files are numbered sequentially, you can use Import(/image sequence.!!
>
> However, the resulting image stack will assume that those slices are
> spaced
> uniformly. If the nonuniform z-position is somewhere encoded in the file
> name or even
> in a binary or text field of each file, you would need to write a (simple
> but) special plugin AFAIK.
>
> Mit freundlichen Grüßen / Best regards
>
> Joachim Wesner
>
>
>
>
> Luciano Paulino
> da Silva
> <lucianopaulinosi An
> [hidden email]> [hidden email]
> Gesendet von: Kopie
> ImageJ Interest
> Group Thema
> <[hidden email]. 3D Display
> GOV>
>
>
> 12.03.2008 14:57
>
>
> Bitte antworten
> an
> ImageJ Interest
> Group
> <[hidden email].
> GOV>
>
>
>
>
>
>
> Dear All,
> We have some 3D surface image files that are not obtained from stacking
> procedures (Atomic force microscopy images - Topography).
> In this way, we have matrix files (512x512; rows x columns) for which the
> corresponding number on the matrix is the height (z axis). How could we
> open such files using ImageJ in order to obtain real surface plots? Are
> there some plugin?
> Luciano and Rebecca
>
>
>
> ______________________________________________________________________
> This email has been scanned by the MessageLabs Email Security System.
> For more information please visit http://www.messagelabs.com/email
> ______________________________________________________________________
>

--
Luciano

Harry Parker

Re: 3D Display

ImageJ can probably open the files directly without a plugin. That depends on the format of your data file, but ImageJ has many builtin options. The options in the File->Import menu are especially powerful for opening unusual data formats.

Many plugins exist for viewing such data in ImageJ. I especially like the plugin "Interactive 3D Surface Plot", listed on ImageJ's Plugin page.

--
Harry Parker
Senior Systems Engineer
Digital Imaging Systems, Inc.

----- Original Message ----
From: Luciano Paulino da Silva <[hidden email]>
To: [hidden email]
Sent: Wednesday, March 12, 2008 9:57:27 AM
Subject: 3D Display

Dear All,
We have some 3D surface image files that are not obtained from stacking
procedures (Atomic force microscopy images - Topography).
In this way, we have matrix files (512x512; rows x columns) for which the
corresponding number on the matrix is the height (z axis). How could we
open such files using ImageJ in order to obtain real surface plots? Are
there some plugin?
Luciano and Rebecca

____________________________________________________________________________________
Never miss a thing. Make Yahoo your home page.
http://www.yahoo.com/r/hs

Joachim Wesner

Antwort: Re: 3D Display

In reply to this post by Luciano Paulino da Silva

I see, so it are not slices, bit a single(?) surface/height info z(x,y),
right? (or there can be several z-entries for the same (x,y)?)

I´m not really familiar with 3D rendering in ImageJ, I know there are some
plugins to do it but cannot suggest the right plugin to treat
such data. Someboy else will certainly chime in......

Mit freundlichen Grüßen / Best regards

Joachim Wesner

Luciano Silva
<lucianopaulinosi
[hidden email]> An
Gesendet von: [hidden email]
ImageJ Interest Kopie
Group
<[hidden email]. Thema
GOV> Re: 3D Display

12.03.2008 15:49

Bitte antworten
an
ImageJ Interest
Group
<[hidden email].
GOV>

Dear Joachim Wesner,
Thank you very much for your suggestions.
We do not have stacked images but only real surface image. In this case, we
just have one image exhibiting all X, Y and Z points in order to be
rendered. In fact, the z-positions are nonuniform. We did not understand
what means the plugin AFAIK. Could you suggest something to do with this
problem?
Could we send you some of our images and the corresponding .txt file of
such
data?
Thanks in advance,
Luciano and Rebecca

2008/3/12, Joachim Wesner <[hidden email]>:

--
Luciano

______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________

Luciano Paulino da Silva

Re: 3D Display

Exactly!
we just have a surface-height image.
Thank you,
Luciano

2008/3/12, Joachim Wesner <[hidden email]>:

>
> I see, so it are not slices, bit a single(?) surface/height info z(x,y),
> right? (or there can be several z-entries for the same (x,y)?)
>
> I´m not really familiar with 3D rendering in ImageJ, I know there are some
> plugins to do it but cannot suggest the right plugin to treat
> such data. Someboy else will certainly chime in......
>
>
> Mit freundlichen Grüßen / Best regards
>
> Joachim Wesner
>
>
>
>
> Luciano Silva
> <lucianopaulinosi
> [hidden email]> An
> Gesendet von: [hidden email]
> ImageJ Interest Kopie
> Group
> <[hidden email]. Thema
> GOV> Re: 3D Display
>
>
> 12.03.2008 15:49
>
>
> Bitte antworten
> an
> ImageJ Interest
> Group
> <[hidden email].
> GOV>
>
>
>
>
>
>
> Dear Joachim Wesner,
> Thank you very much for your suggestions.
> We do not have stacked images but only real surface image. In this case,
> we
> just have one image exhibiting all X, Y and Z points in order to be
> rendered. In fact, the z-positions are nonuniform. We did not understand
> what means the plugin AFAIK. Could you suggest something to do with this
> problem?
> Could we send you some of our images and the corresponding .txt file of
> such
> data?
> Thanks in advance,
> Luciano and Rebecca
>
> 2008/3/12, Joachim Wesner <[hidden email]>:
> >
> > Hi,
> >
> > You mean you have a whole lot of files, each for one z-position? If the
> > files are numbered sequentially, you can use Import(/image sequence.!!
> >
> > However, the resulting image stack will assume that those slices are
> > spaced
> > uniformly. If the nonuniform z-position is somewhere encoded in the file
> > name or even
> > in a binary or text field of each file, you would need to write a
> (simple
> > but) special plugin AFAIK.
> >
> > Mit freundlichen Grüßen / Best regards
> >
> > Joachim Wesner
> >
> >
> >
> >
> > Luciano Paulino
> > da Silva
> >
> <lucianopaulinosi An
> > [hidden email]> [hidden email]
> > Gesendet von:
> Kopie
> > ImageJ Interest
> > Group
> Thema
> > <[hidden email]. 3D Display
> > GOV>
> >
> >
> > 12.03.2008 14:57
> >
> >
> > Bitte antworten
> > an
> > ImageJ Interest
> > Group
> > <[hidden email].
> > GOV>
> >
> >
> >
> >
> >
> >
> > Dear All,
> > We have some 3D surface image files that are not obtained from stacking
> > procedures (Atomic force microscopy images - Topography).
> > In this way, we have matrix files (512x512; rows x columns) for which
> the
> > corresponding number on the matrix is the height (z axis). How could we
> > open such files using ImageJ in order to obtain real surface plots? Are
> > there some plugin?
> > Luciano and Rebecca
> >
> >
> >
> > ______________________________________________________________________
> > This email has been scanned by the MessageLabs Email Security System.
> > For more information please visit http://www.messagelabs.com/email
> > ______________________________________________________________________
> >
>
>
>
> --
> Luciano
>
>
>
> ______________________________________________________________________
> This email has been scanned by the MessageLabs Email Security System.
> For more information please visit http://www.messagelabs.com/email
> ______________________________________________________________________
>

--
Luciano

Leon Espinosa

Re: 3D Display

In reply to this post by Luciano Paulino da Silva

Dear Luciano "surfaceJ" is what you are looking for:

(Michael Abramoff plugins collection)

http://webscreen.ophth.uiowa.edu/bij/surfacej.htm

Cheers, Leon

Le 12 mars 08 à 14:57, Luciano Paulino da Silva a écrit :

Leon Espinosa
Laboratoire des Rickettsies du Pr. RAOULT
UMR CNRS 6020
Fac. de Medecine de la Timone
27 Bd Jean Moulin
13005 Marseille

tel 04 91 38 55 17
fax 04 91 38 77 72

portable: 06 79 25 97 40

Luciano Paulino da Silva

Re: 3D Display

Dear Leon,
In our situation, we have the 3D coordinates in order to be plotted aiming
future segmentation of 3D objects distributed on the surface. We are not
looking for a plugin that is able to render 3D images using color intensity.
Our data are already with X, Y, and Z directions of the surface. We have
these data as a matrix X/Y of defined dimension and for which any value is
the Z (height). In this way, your suggestion is not aimed to solve this.
Thanks in advance,
Luciano

2008/3/12, Leon Espinosa <[hidden email]>:

>
> Dear Luciano "surfaceJ" is what you are looking for:
>
> (Michael Abramoff plugins collection)
>
> http://webscreen.ophth.uiowa.edu/bij/surfacej.htm
>
> Cheers, Leon
>
>
>
> Le 12 mars 08 à 14:57, Luciano Paulino da Silva a écrit :
>
> > Dear All,
> > We have some 3D surface image files that are not obtained from
> > stacking
> > procedures (Atomic force microscopy images - Topography).
> > In this way, we have matrix files (512x512; rows x columns) for
> > which the
> > corresponding number on the matrix is the height (z axis). How could
> > we
> > open such files using ImageJ in order to obtain real surface plots?
> > Are
> > there some plugin?
> > Luciano and Rebecca
> >
>
> Leon Espinosa
> Laboratoire des Rickettsies du Pr. RAOULT
> UMR CNRS 6020
> Fac. de Medecine de la Timone
> 27 Bd Jean Moulin
> 13005 Marseille
>
> tel 04 91 38 55 17
> fax 04 91 38 77 72
>
> portable: 06 79 25 97 40
>

--
Luciano

David Hovis

Re: 3D Display

On Mar 12, 2008, at 12:16 PM, Luciano Silva wrote:

> Dear Leon,
> In our situation, we have the 3D coordinates in order to be plotted
> aiming
> future segmentation of 3D objects distributed on the surface. We are
> not
> looking for a plugin that is able to render 3D images using color
> intensity.
> Our data are already with X, Y, and Z directions of the surface. We
> have
> these data as a matrix X/Y of defined dimension and for which any
> value is
> the Z (height). In this way, your suggestion is not aimed to solve
> this.
> Thanks in advance,
> Luciano

I'm going to repeat myself:

File -> Import -> Text Image...

--David

Luciano Paulino da Silva

Re: 3D Display

Dear David,
I have tried your suggestion but I did not know what is the format to import
a text image. Should we separte each height value using commas or spaces?
Our matrix has 512x512 lines x columns but the reference value to be used
on further calculations is different from this pixels values (eg. a 512x512
matrix could be truth a dimension of 100 µmx100 µm). The height is already
in µm units.
Thanks in advance,
Luciano

2008/3/12, David Hovis <[hidden email]>:

>
> On Mar 12, 2008, at 12:16 PM, Luciano Silva wrote:
>
> > Dear Leon,
> > In our situation, we have the 3D coordinates in order to be plotted
> > aiming
> > future segmentation of 3D objects distributed on the surface. We are
> > not
> > looking for a plugin that is able to render 3D images using color
> > intensity.
> > Our data are already with X, Y, and Z directions of the surface. We
> > have
> > these data as a matrix X/Y of defined dimension and for which any
> > value is
> > the Z (height). In this way, your suggestion is not aimed to solve
> > this.
> > Thanks in advance,
> > Luciano
>
> I'm going to repeat myself:
>
> File -> Import -> Text Image...
>
> --David
>

--
Luciano

David Hovis

Re: 3D Display

On Mar 12, 2008, at 12:45 PM, Luciano Silva wrote:

> Dear David,
> I have tried your suggestion but I did not know what is the format
> to import
> a text image. Should we separte each height value using commas or
> spaces?
> Our matrix has 512x512 lines x columns but the reference value to
> be used
> on further calculations is different from this pixels values (eg. a
> 512x512
> matrix could be truth a dimension of 100 µmx100 µm). The height is
> already
> in µm units.
> Thanks in advance,
> Luciano

I just tested it and it looks like comma separated does not work, but
space or tab separated does. Each line must contain the same number
of columns as the first. The text file should contain only the image
data. Your height data will be preserved in whatever units they are
currently in.

After you load the image, you can set the x-y scale using Image-
>Properties... or Analyze->Set Scale...

--David

Joachim Wesner

Antwort: Re: 3D Display

In reply to this post by David Hovis

Hi Luciano, what David is refering to is that "text image" will take your
single file and create a 2D image where intensity (grey) value represent
height (Step 1), starting from that you likely
can use SurfaceJ to create a true 3D representation of your object(s).
(Step 2)

However, as I mentioned in that other post, if one considers a simple fully
convex closed sphere/cell, one will even in the simplest case get TWO z
values for each x and y inside the projection of that
cell (upper and lower surface), the more folded the surface the more
nonunique entries (x,y). Then text import will not work!? Is that the case
with your data?

Mit freundlichen Grüßen / Best regards

Joachim Wesner

David Hovis
<[hidden email]>
Gesendet von: An
ImageJ Interest [hidden email]
Group Kopie
<[hidden email].
GOV> Thema
Re: 3D Display

12.03.2008 17:22

Bitte antworten
an
ImageJ Interest
Group
<[hidden email].
GOV>

On Mar 12, 2008, at 12:16 PM, Luciano Silva wrote:

I'm going to repeat myself:

File -> Import -> Text Image...

--David

______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________

Luciano Paulino da Silva

Re: 3D Display

Exactly Joachim, our images have an irregular surface to be measured the
volume after segmentation of the background. And in addition to that we have
always more than one unique cell per area, and thses cells must be baseline
filtered prior calculations of volume or other analyses of measurable
parameters. We can now see the image, but we can not display the 3D volume
or measure it...
Luciano

2008/3/12, Joachim Wesner <[hidden email]>:

>
> Hi Luciano, what David is refering to is that "text image" will take your
> single file and create a 2D image where intensity (grey) value represent
> height (Step 1), starting from that you likely
> can use SurfaceJ to create a true 3D representation of your object(s).
> (Step 2)
>
> However, as I mentioned in that other post, if one considers a simple
> fully
> convex closed sphere/cell, one will even in the simplest case get TWO z
> values for each x and y inside the projection of that
> cell (upper and lower surface), the more folded the surface the more
> nonunique entries (x,y). Then text import will not work!? Is that the case
> with your data?
>
> Mit freundlichen Grüßen / Best regards
>
> Joachim Wesner
>
>
>
>
> David Hovis
> <[hidden email]>
> Gesendet von: An
> ImageJ Interest [hidden email]
> Group Kopie
> <[hidden email].
> GOV> Thema
> Re: 3D Display
>
> 12.03.2008 17:22
>
>
> Bitte antworten
> an
> ImageJ Interest
> Group
> <[hidden email].
> GOV>
>
>
>
>
>
>
> On Mar 12, 2008, at 12:16 PM, Luciano Silva wrote:
>
> > Dear Leon,
> > In our situation, we have the 3D coordinates in order to be plotted
> > aiming
> > future segmentation of 3D objects distributed on the surface. We are
> > not
> > looking for a plugin that is able to render 3D images using color
> > intensity.
> > Our data are already with X, Y, and Z directions of the surface. We
> > have
> > these data as a matrix X/Y of defined dimension and for which any
> > value is
> > the Z (height). In this way, your suggestion is not aimed to solve
> > this.
> > Thanks in advance,
> > Luciano
>
> I'm going to repeat myself:
>
> File -> Import -> Text Image...
>
> --David
>
>
>
> ______________________________________________________________________
> This email has been scanned by the MessageLabs Email Security System.
> For more information please visit http://www.messagelabs.com/email
> ______________________________________________________________________
>

--
Luciano

Luciano Paulino da Silva

Re: 3D Display

In reply to this post by David Hovis

Dear David,
We have used TAB text and we could now open the imageb and adjust the
dimensions... Our next step is to measure the volume of individual objects
in such surface. Have you any idea how could we segment these particles?
They are irregular in shape (some of them biconcave and others "hat-like".)
Thanks in advance,
Luciano

2008/3/12, David Hovis <[hidden email]>:

>
> On Mar 12, 2008, at 12:45 PM, Luciano Silva wrote:
>
> > Dear David,
> > I have tried your suggestion but I did not know what is the format
> > to import
> > a text image. Should we separte each height value using commas or
> > spaces?
> > Our matrix has 512x512 lines x columns but the reference value to
> > be used
> > on further calculations is different from this pixels values (eg. a
> > 512x512
> > matrix could be truth a dimension of 100 µmx100 µm). The height is
> > already
> > in µm units.
> > Thanks in advance,
> > Luciano
>
>
> I just tested it and it looks like comma separated does not work, but
> space or tab separated does. Each line must contain the same number
> of columns as the first. The text file should contain only the image
> data. Your height data will be preserved in whatever units they are
> currently in.
>
> After you load the image, you can set the x-y scale using Image-
> >Properties... or Analyze->Set Scale...
>
> --David

--
Luciano

GCH-2

Re: Antwort: Re: 3D Display

In reply to this post by Joachim Wesner

Hi,

If I remember correctly, the problem with some AFM images is that
they may be composed of several channels. If your images are ascii,
512x512 pixels the File/Import/Text image... shuodl do the job.
Somewhere I also have a macro (provided mostly by Wayne) that I used
to open AFM images (single or series within a folder) automatically.
I can send it to you in a separate email if necessary.

Once you open the images you can use the Interactive 3D surface plot
by Kai Uwe Barthel (http://rsb.info.nih.gov/ij/plugins/surface-
plot-3d.html) to render and visualize the surface. It works great!

You can also use the SurfCharJ to characterize and describe the
surface topography.
http://www.gcsca.net/IJ/SurfCharJ.html

I hope this helps,

Gary.
http://www.gcsca.net

On Mar 12, 2008, at 6:02 PM, Joachim Wesner wrote:

> Hi Luciano, what David is refering to is that "text image" will
> take your
> single file and create a 2D image where intensity (grey) value
> represent
> height (Step 1), starting from that you likely
> can use SurfaceJ to create a true 3D representation of your object(s).
> (Step 2)
>
> However, as I mentioned in that other post, if one considers a
> simple fully
> convex closed sphere/cell, one will even in the simplest case get
> TWO z
> values for each x and y inside the projection of that
> cell (upper and lower surface), the more folded the surface the more
> nonunique entries (x,y). Then text import will not work!? Is that
> the case
> with your data?
>
> Mit freundlichen Grüßen / Best regards
>
> Joachim Wesner
>
>
>
>
> David Hovis
> <[hidden email]>
> Gesendet
> von: An
> ImageJ Interest [hidden email]
>
> Group Kopie
> <[hidden email].
>
> GOV> Thema
> Re: 3D Display
>
> 12.03.2008 17:22
>
>
> Bitte antworten
> an
> ImageJ Interest
> Group
> <[hidden email].
> GOV>
>
>
>
>
>
>
> On Mar 12, 2008, at 12:16 PM, Luciano Silva wrote:
>
>> Dear Leon,
>> In our situation, we have the 3D coordinates in order to be plotted
>> aiming
>> future segmentation of 3D objects distributed on the surface. We are
>> not
>> looking for a plugin that is able to render 3D images using color
>> intensity.
>> Our data are already with X, Y, and Z directions of the surface. We
>> have
>> these data as a matrix X/Y of defined dimension and for which any
>> value is
>> the Z (height). In this way, your suggestion is not aimed to solve
>> this.
>> Thanks in advance,
>> Luciano
>
> I'm going to repeat myself:
>
> File -> Import -> Text Image...
>
> --David
>
>
>
> ______________________________________________________________________
> This email has been scanned by the MessageLabs Email Security System.
> For more information please visit http://www.messagelabs.com/email
> ______________________________________________________________________
>
>

Christopher Yip

Re: Antwort: Re: 3D Display

Gary's right - depends on the vendor but sometimes the AFM data set
has an ASCII header with parameters in it, a buffer and then the data
sets in some format (I'm used to the DI / Veeco format).

There have been lots of macros / plug-ins written over the years to
import AFM data of various forms from various vendors.

One thing to be careful about - esp if you are measuring volumes - the
tip convolution effect can / does make things much wider than expected
(depending on what you are looking at) so uncorrected measurements of
volume can be problematic...

Chris

On Mar 12, 2008, at 1:40 PM, GCH wrote:

> Hi,
>
> If I remember correctly, the problem with some AFM images is that
> they may be composed of several channels. If your images are ascii,
> 512x512 pixels the File/Import/Text image... shuodl do the job.
> Somewhere I also have a macro (provided mostly by Wayne) that I used
> to open AFM images (single or series within a folder)
> automatically. I can send it to you in a separate email if necessary.
>
> Once you open the images you can use the Interactive 3D surface plot
> by Kai Uwe Barthel (http://rsb.info.nih.gov/ij/plugins/surface-plot-3d.html
> ) to render and visualize the surface. It works great!
>
> You can also use the SurfCharJ to characterize and describe the
> surface topography.
> http://www.gcsca.net/IJ/SurfCharJ.html
>
> I hope this helps,
>
> Gary.
> http://www.gcsca.net

Christopher M. Yip Ph.D., P.Eng
[hidden email]
http://bigten.med.utoronto.ca

Luciano Paulino da Silva

Re: Antwort: Re: 3D Display

In reply to this post by GCH-2

Dear Gary,
The images obtaines of our instrument could not be directly opened using any
plugin devoted for AFM. We have a SPM-9600 Shimadzu. However, we can export
as an image file. In this way, we had success to open and draw the 3D image
using your suggestion (Interactive 3D surface plot). Thank you very much!
At this moment, we are trying to segment objects in such image in order to
obtain the absolute volume of each object dispersed in the surface. For each
image we can see dozens of the irregular particles we want measure. Have you
some suggestion to perform this?
Thanks in advance,
Luciano

2008/3/12, GCH <[hidden email]>:

>
> Hi,
>
> If I remember correctly, the problem with some AFM images is that
> they may be composed of several channels. If your images are ascii,
> 512x512 pixels the File/Import/Text image... shuodl do the job.
> Somewhere I also have a macro (provided mostly by Wayne) that I used
> to open AFM images (single or series within a folder) automatically.
> I can send it to you in a separate email if necessary.
>
> Once you open the images you can use the Interactive 3D surface plot
> by Kai Uwe Barthel (http://rsb.info.nih.gov/ij/plugins/surface-
> plot-3d.html) to render and visualize the surface. It works great!
>
> You can also use the SurfCharJ to characterize and describe the
> surface topography.
> http://www.gcsca.net/IJ/SurfCharJ.html
>
> I hope this helps,
>
> Gary.
> http://www.gcsca.net
>
>
>
>
>
> On Mar 12, 2008, at 6:02 PM, Joachim Wesner wrote:
>
> > Hi Luciano, what David is refering to is that "text image" will
> > take your
> > single file and create a 2D image where intensity (grey) value
> > represent
> > height (Step 1), starting from that you likely
> > can use SurfaceJ to create a true 3D representation of your object(s).
> > (Step 2)
> >
> > However, as I mentioned in that other post, if one considers a
> > simple fully
> > convex closed sphere/cell, one will even in the simplest case get
> > TWO z
> > values for each x and y inside the projection of that
> > cell (upper and lower surface), the more folded the surface the more
> > nonunique entries (x,y). Then text import will not work!? Is that
> > the case
> > with your data?
> >
> > Mit freundlichen Grüßen / Best regards
> >
> > Joachim Wesner
> >
> >
> >
> >
> > David Hovis
> > <[hidden email]>
> > Gesendet
> > von: An
> > ImageJ Interest [hidden email]
> >
> > Group Kopie
> > <[hidden email].
> >
> > GOV> Thema
> > Re: 3D Display
> >
> > 12.03.2008 17:22
> >
> >
> > Bitte antworten
> > an
> > ImageJ Interest
> > Group
> > <[hidden email].
> > GOV>
> >
> >
> >
> >
> >
> >
> > On Mar 12, 2008, at 12:16 PM, Luciano Silva wrote:
> >
> >> Dear Leon,
> >> In our situation, we have the 3D coordinates in order to be plotted
> >> aiming
> >> future segmentation of 3D objects distributed on the surface. We are
> >> not
> >> looking for a plugin that is able to render 3D images using color
> >> intensity.
> >> Our data are already with X, Y, and Z directions of the surface. We
> >> have
> >> these data as a matrix X/Y of defined dimension and for which any
> >> value is
> >> the Z (height). In this way, your suggestion is not aimed to solve
> >> this.
> >> Thanks in advance,
> >> Luciano
> >
> > I'm going to repeat myself:
> >
> > File -> Import -> Text Image...
> >
> > --David
> >
> >
> >
> > ______________________________________________________________________
> > This email has been scanned by the MessageLabs Email Security System.
> > For more information please visit http://www.messagelabs.com/email
> > ______________________________________________________________________
> >
> >
>

--
Luciano

Luciano Paulino da Silva

Re: Antwort: Re: 3D Display

In reply to this post by Christopher Yip

Dear Christopher,
Thank you very much for your suggestion... In fact, we have not found any
plugin able to directly convert SHIMADZU files into some recognized format
of ImageJ. However, we could export for text image file.
You are right regarding the volumes. The tip convolution effect is highly
observed when analyzing AFM data, for example volumes. However, in our case
we have micrometer objects and this effect is almost always minimized when
compared to nanometric investigation. Our major current problem is related
to an efficient strategy of 3D segmentation. We have tried use half-height
segmentation but id do not function well for our objects (particles) since
they are not regular (some of them resembles to a hat).
Have you some idea of efficient programs to segment at the baselineof each
object from an imagen with dozens of these particles?
Thanks in advance,
Luciano

2008/3/12, Christopher Yip <[hidden email]>:

>
> Gary's right - depends on the vendor but sometimes the AFM data set
> has an ASCII header with parameters in it, a buffer and then the data
> sets in some format (I'm used to the DI / Veeco format).
>
> There have been lots of macros / plug-ins written over the years to
> import AFM data of various forms from various vendors.
>
> One thing to be careful about - esp if you are measuring volumes - the
> tip convolution effect can / does make things much wider than expected
> (depending on what you are looking at) so uncorrected measurements of
> volume can be problematic...
>
> Chris
>
>
> On Mar 12, 2008, at 1:40 PM, GCH wrote:
>
> > Hi,
> >
> > If I remember correctly, the problem with some AFM images is that
> > they may be composed of several channels. If your images are ascii,
> > 512x512 pixels the File/Import/Text image... shuodl do the job.
> > Somewhere I also have a macro (provided mostly by Wayne) that I used
> > to open AFM images (single or series within a folder)
> > automatically. I can send it to you in a separate email if necessary.
> >
> > Once you open the images you can use the Interactive 3D surface plot
> > by Kai Uwe Barthel (
> http://rsb.info.nih.gov/ij/plugins/surface-plot-3d.html
> > ) to render and visualize the surface. It works great!
> >
> > You can also use the SurfCharJ to characterize and describe the
> > surface topography.
> > http://www.gcsca.net/IJ/SurfCharJ.html
> >
> > I hope this helps,
> >
> > Gary.
> > http://www.gcsca.net
>
> Christopher M. Yip Ph.D., P.Eng
> [hidden email]
> http://bigten.med.utoronto.ca
>

--
Luciano

GCH-2

Re: Antwort: Re: 3D Display

In reply to this post by Luciano Paulino da Silva

Luciano,

Can you post or send an image directly to my account for
exemplification?

Gary.
http://www.gcsca.net

On Mar 12, 2008, at 6:56 PM, Luciano Silva wrote:

> Dear Gary,
> The images obtaines of our instrument could not be directly opened
> using any
> plugin devoted for AFM. We have a SPM-9600 Shimadzu. However, we
> can export
> as an image file. In this way, we had success to open and draw the
> 3D image
> using your suggestion (Interactive 3D surface plot). Thank you very
> much!
> At this moment, we are trying to segment objects in such image in
> order to
> obtain the absolute volume of each object dispersed in the surface.
> For each
> image we can see dozens of the irregular particles we want measure.
> Have you
> some suggestion to perform this?
> Thanks in advance,
> Luciano
>
>
> 2008/3/12, GCH <[hidden email]>:
>>
>> Hi,
>>
>> If I remember correctly, the problem with some AFM images is that
>> they may be composed of several channels. If your images are ascii,
>> 512x512 pixels the File/Import/Text image... shuodl do the job.
>> Somewhere I also have a macro (provided mostly by Wayne) that I used
>> to open AFM images (single or series within a folder) automatically.
>> I can send it to you in a separate email if necessary.
>>
>> Once you open the images you can use the Interactive 3D surface plot
>> by Kai Uwe Barthel (http://rsb.info.nih.gov/ij/plugins/surface-
>> plot-3d.html) to render and visualize the surface. It works great!
>>
>> You can also use the SurfCharJ to characterize and describe the
>> surface topography.
>> http://www.gcsca.net/IJ/SurfCharJ.html
>>
>> I hope this helps,
>>
>> Gary.
>> http://www.gcsca.net
>>
>>
>>
>>
>>
>> On Mar 12, 2008, at 6:02 PM, Joachim Wesner wrote:
>>
>>> Hi Luciano, what David is refering to is that "text image" will
>>> take your
>>> single file and create a 2D image where intensity (grey) value
>>> represent
>>> height (Step 1), starting from that you likely
>>> can use SurfaceJ to create a true 3D representation of your object
>>> (s).
>>> (Step 2)
>>>
>>> However, as I mentioned in that other post, if one considers a
>>> simple fully
>>> convex closed sphere/cell, one will even in the simplest case get
>>> TWO z
>>> values for each x and y inside the projection of that
>>> cell (upper and lower surface), the more folded the surface the more
>>> nonunique entries (x,y). Then text import will not work!? Is that
>>> the case
>>> with your data?
>>>
>>> Mit freundlichen Grüßen / Best regards
>>>
>>> Joachim Wesner
>>>
>>>
>>>
>>>
>>> David Hovis
>>> <[hidden email]>
>>> Gesendet
>>> von: An
>>> ImageJ Interest [hidden email]
>>>
>>> Group Kopie
>>> <[hidden email].
>>>
>>> GOV> Thema
>>> Re: 3D Display
>>>
>>> 12.03.2008 17:22
>>>
>>>
>>> Bitte antworten
>>> an
>>> ImageJ Interest
>>> Group
>>> <[hidden email].
>>> GOV>
>>>
>>>
>>>
>>>
>>>
>>>
>>> On Mar 12, 2008, at 12:16 PM, Luciano Silva wrote:
>>>
>>>> Dear Leon,
>>>> In our situation, we have the 3D coordinates in order to be plotted
>>>> aiming
>>>> future segmentation of 3D objects distributed on the surface. We
>>>> are
>>>> not
>>>> looking for a plugin that is able to render 3D images using color
>>>> intensity.
>>>> Our data are already with X, Y, and Z directions of the surface. We
>>>> have
>>>> these data as a matrix X/Y of defined dimension and for which any
>>>> value is
>>>> the Z (height). In this way, your suggestion is not aimed to solve
>>>> this.
>>>> Thanks in advance,
>>>> Luciano
>>>
>>> I'm going to repeat myself:
>>>
>>> File -> Import -> Text Image...
>>>
>>> --David
>>>
>>>
>>>
>>> ____________________________________________________________________
>>> __
>>> This email has been scanned by the MessageLabs Email Security
>>> System.
>>> For more information please visit http://www.messagelabs.com/email
>>> ____________________________________________________________________
>>> __
>>>
>>>
>>
>
>
>
> --
> Luciano