ImageJ

distance between adjacent particles

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

18 messages Options

Girault France

distance between adjacent particles

Dear all,
I have an image with many particles, distributed randomly. I'm trying to get the average distance between a particle with all its closest neighbors but not all of the particles on the image.
Is there an easy way to do or an existing plugin that allows such measurement?

Many thanks,

France
France E. Girault
Geological Institute
CAB E-62
ETH Zürich
Universitätstrasse 6
8092 Zürich
Private phone: ++4178/726.99.20
Office phone: ++4144/632 84 16
Fax: +4144/632 10 80

[hidden email] <mailto:[hidden email]>
http://www.geology.ethz.ch/people/PhDStudents/giraultf/index <http://www.geology.ethz.ch/people/PhDStudents/giraultf/index>

Michael Schmid

Re: distance between adjacent particles

Hi France,

one method that I sometimes use is the following:
- Create one single point per particle, e.g. using
Process>Binary>FindMaxima.
- Run an Autocorrelation (Process>FFT>FT Math correlate, do inverse)
- Use the Radial Profile Plot
http://rsb.info.nih.gov/ij/plugins/radial-profile.html

This gives you the radial distribution function.

Ignore the peak at zero distance.
There will be also a minimum next to zero distance because you
cannot have two isolated points touching each other. This
tells you nothing important about the particle distribution.

If the particles are really randomly distributed, apart from
these two effect and statistical noise, the radial distribution
function should be flat.

Michael
________________________________________________________________

On 9 May 2007, at 22:12, Girault France wrote:

> Dear all,
> I have an image with many particles, distributed randomly. I'm
> trying to get the average distance between a particle with all its
> closest neighbors but not all of the particles on the image.
> Is there an easy way to do or an existing plugin that allows such
> measurement?
>
> Many thanks,
>
> France

karo03

Re: distance between adjacent particles

In reply to this post by Girault France

Typically the neighbors of points under a Delaunay triangulation are
what can be considered as the 'closest neighbors'. There is a plugin
to generate the Delaunay triangulation from selected points (http://
wbgn013.biozentrum.uni-wuerzburg.de/ImageJ/delaunay.html). Possibly
the author knows how to extract the neighbors for each point
including the distance! from the plugin. In fact it would be a nice
extension to graph operations in ImageJ.

Regards
Karsten

Am 09.05.2007 um 22:12 schrieb Girault France:

>
> Dear all,
> I have an image with many particles, distributed randomly. I'm
> trying to get the average distance between a particle with all its
> closest neighbors but not all of the particles on the image.
> Is there an easy way to do or an existing plugin that allows such
> measurement?
>
> Many thanks,
>
> France
> France E. Girault
> Geological Institute
> CAB E-62
> ETH Zürich
> Universitätstrasse 6
> 8092 Zürich
> Private phone: ++4178/726.99.20
> Office phone: ++4144/632 84 16
> Fax: +4144/632 10 80
>
> [hidden email] <mailto:[hidden email]>
> http://www.geology.ethz.ch/people/PhDStudents/giraultf/index
> <http://www.geology.ethz.ch/people/PhDStudents/giraultf/index>
>

Gabriel Landini

Re: distance between adjacent particles

In reply to this post by Michael Schmid

On Wednesday 09 May 2007 21:30:10 Michael Schmid wrote:
> > I have an image with many particles, distributed randomly. I'm
> > trying to get the average distance between a particle with all its
> > closest neighbors but not all of the particles on the image.
> > Is there an easy way to do or an existing plugin that allows such
> > measurement?

To do this using brute-force would be quite easy with a macro:

get the coordinates of all the points or centroids in arrays x() and y(),
for each point calculate using pythagoras, the distance to all other points
and keep the shortest one in another array.
Now look at the distribution of shortest distances.

Cheers,

G.

Gabriel Landini

Re: distance between adjacent particles

In reply to this post by karo03

On Wednesday 09 May 2007 22:16:41 Karsten Rodenacker wrote:
> Typically the neighbors of points under a Delaunay triangulation are
> what can be considered as the 'closest neighbors'.

You are right, the nearest-neighbour graph is a subgraph of the delaunay
triangulation, but I think it is easier to calculate the nearest neighbours
in the way I mentioned earlier because for the delaunay graph one has to get
the points and distances to all the other points anyway.

Regards,

Gabriel

William Guilford

Re: distance between adjacent particles

In reply to this post by Girault France

An alternative to a strict "nearest neighbor" is to generate the pair-wise
distances from each centroid to every other centroid in the image. Compile
these into a "pair-wise difference histogram" of distances. This is the
equivalent of a 1D autocorrelation function. You will see a peak at the mean
nearest neighbor distance.

~ Will

-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
Girault France
Sent: Wednesday, May 09, 2007 4:13 PM
To: [hidden email]
Subject: distance between adjacent particles

Dear all,
I have an image with many particles, distributed randomly. I'm trying to get
the average distance between a particle with all its closest neighbors but
not all of the particles on the image.
Is there an easy way to do or an existing plugin that allows such
measurement?

Many thanks,

France
France E. Girault
Geological Institute
CAB E-62
ETH Zürich
Universitätstrasse 6
8092 Zürich
Private phone: ++4178/726.99.20
Office phone: ++4144/632 84 16
Fax: +4144/632 10 80

[hidden email] <mailto:[hidden email]>
http://www.geology.ethz.ch/people/PhDStudents/giraultf/index
<http://www.geology.ethz.ch/people/PhDStudents/giraultf/index>

karo03

Re: distance between adjacent particles

In reply to this post by Gabriel Landini

Hi, Gabriel
Just for clarification: The set of closest neighbored points is
something more than NNG (nearest-neighbour graph). It has something
to do with our visual behavior, how we relate neighbored objects. One
result of DT (Delaunay triangulation) could be the list of neighbors
per point. From which easily an 'arrangement parameter' (mean of
distances) could be deduced. This again is of course highly related
to the mean area of Voronoi polygones of one point or the watershed
bassins neighbored to a region.

Regards
Karsten

Am 09.05.2007 um 23:57 schrieb Gabriel Landini:

> On Wednesday 09 May 2007 22:16:41 Karsten Rodenacker wrote:
>> Typically the neighbors of points under a Delaunay triangulation are
>> what can be considered as the 'closest neighbors'.
>
> You are right, the nearest-neighbour graph is a subgraph of the
> delaunay
> triangulation, but I think it is easier to calculate the nearest
> neighbours
> in the way I mentioned earlier because for the delaunay graph one
> has to get
> the points and distances to all the other points anyway.
>
> Regards,
>
> Gabriel

jchanson

Re: distance between adjacent particles

In reply to this post by Gabriel Landini

Ok. If you're not really looking for "closest neighbor" distance
but the "average distance among the closest neighbors"...
You'll need to decide how many of the closest neighbors you want
to include in the average. Depending on what you expect the distribution
might be like, you could use from 3-6 neighbors. Then, you could do the
"brute force" calculation suggested below, except instead of keeping the
shortest distance in an array, you would keep the "n" shortest distances
in an array. When you were done scanning all of the points, you would
find the average of those "n" points. I like using "6" points, because
that's how many neighbors an object would have if it was in a tightly
packed hexagonal pattern.

In other applications, I've been interested finding areas of
different "densities" of objects and I've used the following steps:
- threshold the objects
- reduce the objects to their ultimate center points
- run a large scale blur (scale dependent on how far away you want to
include neighbors)
- this image will have it's highest intensities in the areas where you
have the closest packing of objects
This represents more of a "texture" measure. The advantage of
this technique is that it doesn't require any extra java programming, just
image processing steps. The problem with this technique is that it's a
lot harder to try to describe what the results mean.

good luck,
Jeff

Gabriel Landini <[hidden email]>
Sent by: ImageJ Interest Group <[hidden email]>
05/09/2007 05:36 PM
Please respond to
ImageJ Interest Group <[hidden email]>

To
[hidden email]
cc

Subject
Re: distance between adjacent particles

On Wednesday 09 May 2007 21:30:10 Michael Schmid wrote:
> > I have an image with many particles, distributed randomly. I'm
> > trying to get the average distance between a particle with all its
> > closest neighbors but not all of the particles on the image.
> > Is there an easy way to do or an existing plugin that allows such
> > measurement?

To do this using brute-force would be quite easy with a macro:

get the coordinates of all the points or centroids in arrays x() and y(),
for each point calculate using pythagoras, the distance to all other
points
and keep the shortest one in another array.
Now look at the distribution of shortest distances.

Cheers,

G.

Girault France

Re: distance between adjacent particles

Dear all,
Thanks for all the replies,

Following your advices, I've been trying some techniques to get
distances between particles. Here I try to get the shortest distance
between each particle and its closest neightbor as suggested by G.
Landini. First I have to assess the distance between the particle and
each of the other particle, avoiding the value "0" for the distance:
(see below)

With this macro I still get the value 0 in some cases and I don't know
how to get rid of it. Maybe there also is an easier way to get those
values?

Secondly I just saw the last reply about taking 6 closest neighbors
which would exactly fit with I want to do (4 closest neighbors rather
than 6), but I don't really see how I can get this right now with what
I've done. Any ideas on how to do this? Creating a second array and
store the 4 smallest values and average them at the end?

Many thanks,

France

var dist1;
var dist;
for (i=0; i<nResults; i++) {
XM=getResult("XM", i);
YM=getResult("YM",i);
a=nResults;
a1 = newArray(a);
for (b=0; b<nResults; b++){
XM2=getResult("XM", b);
YM2=getResult("YM",b);
dist1=dist;

dist=sqrt(((XM2-XM)*(XM2-XM))+((YM2-YM)*(YM2-YM)));
a1[b] = dist;
}
for (b=0;b<a;b++){
dist=a1[b];
if(dist==0){dist=10000;}
if (dist>dist1) {dist=dist1;}
dist1=dist;
}
print(dist1);
setResult("shortest_dist", i, dist1);
}

updateResults();

-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
Jeffrey C Hanson
Sent: jeudi, 10. mai 2007 17:13
To: [hidden email]
Subject: Re: distance between adjacent particles

Ok. If you're not really looking for "closest neighbor"
distance
but the "average distance among the closest neighbors"...
You'll need to decide how many of the closest neighbors you want

to include in the average. Depending on what you expect the
distribution
might be like, you could use from 3-6 neighbors. Then, you could do the

"brute force" calculation suggested below, except instead of keeping the

shortest distance in an array, you would keep the "n" shortest distances

in an array. When you were done scanning all of the points, you would
find the average of those "n" points. I like using "6" points, because
that's how many neighbors an object would have if it was in a tightly
packed hexagonal pattern.

In other applications, I've been interested finding areas of
different "densities" of objects and I've used the following steps:
- threshold the objects
- reduce the objects to their ultimate center points
- run a large scale blur (scale dependent on how far away you want to
include neighbors)
- this image will have it's highest intensities in the areas where you
have the closest packing of objects
This represents more of a "texture" measure. The advantage of
this technique is that it doesn't require any extra java programming,
just
image processing steps. The problem with this technique is that it's a
lot harder to try to describe what the results mean.

good luck,
Jeff

Gabriel Landini <[hidden email]>
Sent by: ImageJ Interest Group <[hidden email]>
05/09/2007 05:36 PM
Please respond to
ImageJ Interest Group <[hidden email]>

To
[hidden email]
cc

Subject
Re: distance between adjacent particles

On Wednesday 09 May 2007 21:30:10 Michael Schmid wrote:
> > I have an image with many particles, distributed randomly. I'm
> > trying to get the average distance between a particle with all its
> > closest neighbors but not all of the particles on the image.
> > Is there an easy way to do or an existing plugin that allows such
> > measurement?

To do this using brute-force would be quite easy with a macro:

get the coordinates of all the points or centroids in arrays x() and
y(),
for each point calculate using pythagoras, the distance to all other
points
and keep the shortest one in another array.
Now look at the distribution of shortest distances.

Cheers,

G.

jchanson

Re: distance between adjacent particles

France -
It looks like you initialize the value of "dist1" inside the for
loop where you're calculating all of the distances. You should just
initialize it to a large value outside of the for loop. Try this,
instead:

var dist1;
var dist;
for (i=0; i<nResults; i++) {
XM=getResult("XM", i);
YM=getResult("YM",i);
a=nResults;
a1 = newArray(a);
for (b=0; b<nResults; b++){
XM2=getResult("XM", b);
YM2=getResult("YM",b);
dist=sqrt(((XM2-XM)*(XM2-XM))+((YM2-YM)*(YM2-YM)));
a1[b] = dist;
}
dist1=10000;
for (b=0;b<a;b++){
dist=a1[b];
if(dist==0){dist=10000;}
if (dist>dist1)
{dist=dist1;}
dist1=dist;
}
print(dist1);
setResult("shortest_dist", i, dist1);
}

updateResults();

Jeff
Senior Imaging Analyst
[hidden email]

Gabriel Landini

Re: distance between adjacent particles

On Thursday 10 May 2007 18:48:09 Jeffrey C Hanson wrote:
> It looks like you initialize the value of "dist1" inside the for
> loop where you're calculating all of the distances. You should just
> initialize it to a large value outside of the for loop. Try this,
> instead:
[...]

> dist=sqrt(((XM2-XM)*(XM2-XM))+((YM2-YM)*(YM2-YM)));

A handy trick. You can save some time by not doing the sqrt and comparing the
squares instead.
Then do the sqrt at the end only on the values you kept.
I hope it helps.

Cheers,

G.

Dan Batzel

Re: distance between adjacent particles

In reply to this post by Girault France

Some concepts that may be thought provoking, "moves to crowding" and
"distance to regularity", attributed to J. N. Perry & M. Hewitt (1991)
and R. Alston (1994), are described in a document deposited here:

http://www.rothamsted.ac.uk/pie/sadie/reprints/perry1995a_jae.pdf

Dan Batzel, PhD
Research Scientist
Gentex Corporation
Carbondale, PA

-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
Girault France
Sent: Thursday, May 10, 2007 11:30 AM
To: [hidden email]
Subject: Re: distance between adjacent particles

Dear all,
Thanks for all the replies,

Following your advices, I've been trying some techniques to get
distances between particles. Here I try to get the shortest distance
between each particle and its closest neightbor as suggested by G.
Landini. First I have to assess the distance between the particle and
each of the other particle, avoiding the value "0" for the distance:
(see below)

With this macro I still get the value 0 in some cases and I don't know
how to get rid of it. Maybe there also is an easier way to get those
values?

Secondly I just saw the last reply about taking 6 closest neighbors
which would exactly fit with I want to do (4 closest neighbors rather
than 6), but I don't really see how I can get this right now with what
I've done. Any ideas on how to do this? Creating a second array and
store the 4 smallest values and average them at the end?

Many thanks,

France

var dist1;
var dist;
for (i=0; i<nResults; i++) {
XM=getResult("XM", i);
YM=getResult("YM",i);
a=nResults;
a1 = newArray(a);
for (b=0; b<nResults; b++){
XM2=getResult("XM", b);
YM2=getResult("YM",b);
dist1=dist;

dist=sqrt(((XM2-XM)*(XM2-XM))+((YM2-YM)*(YM2-YM)));
a1[b] = dist;
}
for (b=0;b<a;b++){
dist=a1[b];
if(dist==0){dist=10000;}
if (dist>dist1) {dist=dist1;}
dist1=dist;
}
print(dist1);
setResult("shortest_dist", i, dist1);
}

updateResults();

-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
Jeffrey C Hanson
Sent: jeudi, 10. mai 2007 17:13
To: [hidden email]
Subject: Re: distance between adjacent particles

Ok. If you're not really looking for "closest neighbor"
distance
but the "average distance among the closest neighbors"...
You'll need to decide how many of the closest neighbors you want

to include in the average. Depending on what you expect the
distribution
might be like, you could use from 3-6 neighbors. Then, you could do the

"brute force" calculation suggested below, except instead of keeping the

shortest distance in an array, you would keep the "n" shortest distances

in an array. When you were done scanning all of the points, you would
find the average of those "n" points. I like using "6" points, because
that's how many neighbors an object would have if it was in a tightly
packed hexagonal pattern.

In other applications, I've been interested finding areas of
different "densities" of objects and I've used the following steps:
- threshold the objects
- reduce the objects to their ultimate center points
- run a large scale blur (scale dependent on how far away you want to
include neighbors)
- this image will have it's highest intensities in the areas where you
have the closest packing of objects
This represents more of a "texture" measure. The advantage of
this technique is that it doesn't require any extra java programming,
just
image processing steps. The problem with this technique is that it's a
lot harder to try to describe what the results mean.

good luck,
Jeff

Gabriel Landini <[hidden email]>
Sent by: ImageJ Interest Group <[hidden email]>
05/09/2007 05:36 PM
Please respond to
ImageJ Interest Group <[hidden email]>

To
[hidden email]
cc

Subject
Re: distance between adjacent particles

On Wednesday 09 May 2007 21:30:10 Michael Schmid wrote:
> > I have an image with many particles, distributed randomly. I'm
> > trying to get the average distance between a particle with all its
> > closest neighbors but not all of the particles on the image.
> > Is there an easy way to do or an existing plugin that allows such
> > measurement?

To do this using brute-force would be quite easy with a macro:

get the coordinates of all the points or centroids in arrays x() and
y(),
for each point calculate using pythagoras, the distance to all other
points
and keep the shortest one in another array.
Now look at the distribution of shortest distances.

Cheers,

G.

dobensl

Re: distance between adjacent particles

In reply to this post by jchanson

This is a very useful macro and almost takes me to where I want to get - is there a way to find the values for XM2 and YM2 and add these to the results table? Also, could someone briefly describe the logic behind the notion: "initialize it to a large value outside of the for loop"?

Thanks,

Len Dobens

BenTupper

Re: distance between adjacent particles

Hi Len,

That's quite an old thread in Nabble. I'm not sure if it will help you, but you could try http://rsb.info.nih.gov/ij/plugins/graph/index.html There may be newer tools available that I am not familiar with.

Ben

> On May 22, 2016, at 4:06 PM, dobensl <[hidden email]> wrote:
>
> This is a very useful macro and almost takes me to where I want to get - is
> there a way to find the values for XM2 and YM2 and add these to the results
> table? Also, could someone briefly describe the logic behind the notion:
> "initialize it to a large value outside of the for loop"?
>
> Thanks,
>
> Len Dobens
>
>
>
> --
> View this message in context: http://imagej.1557.x6.nabble.com/distance-between-adjacent-particles-tp3699485p5016492.html
> Sent from the ImageJ mailing list archive at Nabble.com.
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

Cammer, Michael

Re: distance between adjacent particles

Curious if anyone thinks this would be useful. It may only need a few lines of code to list the pairs of points and the distances. I wrote it just because I felt like it but no one has brought me a problem to apply it to.
http://microscopynotes.com/imagej/macros/shortest_distances_and_clustering/index.html

=========================================================================
Michael Cammer, Microscopy Core & Skirball Institute, NYU Langone Medical Center
Cell: 914-309-3270 ** Office: Skirball 2nd Floor main office, back right **
http://ocs.med.nyu.edu/microscopy & http://microscopynotes.com/

-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Ben Tupper
Sent: Monday, May 23, 2016 1:08 PM
To: [hidden email]
Subject: Re: distance between adjacent particles

Hi Len,

That's quite an old thread in Nabble. I'm not sure if it will help you, but you could try https://urldefense.proofpoint.com/v2/url?u=http-3A__rsb.info.nih.gov_ij_plugins_graph_index.html&d=CwIFAg&c=j5oPpO0eBH1iio48DtsedbOBGmuw5jHLjgvtN2r4ehE&r=oU_05LztNstAydlbm5L5GDu_vAdjXk3frDLx_CqKkuo&m=UoPrt3a6dSRFDTVKNKhxQxr9hVUnIlg2qHZVtrhy_1M&s=njt95X9CC5z1_RlTbXErpNZ8qshIp4aaCxkviC1GaJI&e= There may be newer tools available that I am not familiar with.

Ben

> On May 22, 2016, at 4:06 PM, dobensl <[hidden email]> wrote:
>
> This is a very useful macro and almost takes me to where I want to get
> - is there a way to find the values for XM2 and YM2 and add these to
> the results table? Also, could someone briefly describe the logic behind the notion:
> "initialize it to a large value outside of the for loop"?
>
> Thanks,
>
> Len Dobens
>
>
>
> --
> View this message in context:
> https://urldefense.proofpoint.com/v2/url?u=http-3A__imagej.1557.x6.nab
> ble.com_distance-2Dbetween-2Dadjacent-2Dparticles-2Dtp3699485p5016492.
> html&d=CwIFAg&c=j5oPpO0eBH1iio48DtsedbOBGmuw5jHLjgvtN2r4ehE&r=oU_05Lzt
> NstAydlbm5L5GDu_vAdjXk3frDLx_CqKkuo&m=UoPrt3a6dSRFDTVKNKhxQxr9hVUnIlg2
> qHZVtrhy_1M&s=SFCtTsIXuwdtQq58bn7D7GiNgO3u5XcXK5w8iufElds&e=
> Sent from the ImageJ mailing list archive at Nabble.com.
>
> --
> ImageJ mailing list:
> https://urldefense.proofpoint.com/v2/url?u=http-3A__imagej.nih.gov_ij_
> list.html&d=CwIFAg&c=j5oPpO0eBH1iio48DtsedbOBGmuw5jHLjgvtN2r4ehE&r=oU_
> 05LztNstAydlbm5L5GDu_vAdjXk3frDLx_CqKkuo&m=UoPrt3a6dSRFDTVKNKhxQxr9hVU
> nIlg2qHZVtrhy_1M&s=A6W4FamOmzV5A4xjupQZMH0LP7G2kO3EEK6U_jH8mbI&e=

--
ImageJ mailing list: https://urldefense.proofpoint.com/v2/url?u=http-3A__imagej.nih.gov_ij_list.html&d=CwIFAg&c=j5oPpO0eBH1iio48DtsedbOBGmuw5jHLjgvtN2r4ehE&r=oU_05LztNstAydlbm5L5GDu_vAdjXk3frDLx_CqKkuo&m=UoPrt3a6dSRFDTVKNKhxQxr9hVUnIlg2qHZVtrhy_1M&s=A6W4FamOmzV5A4xjupQZMH0LP7G2kO3EEK6U_jH8mbI&e=

------------------------------------------------------------
This email message, including any attachments, is for the sole use of the intended recipient(s) and may contain information that is proprietary, confidential, and exempt from disclosure under applicable law. Any unauthorized review, use, disclosure, or distribution is prohibited. If you have received this email in error please notify the sender by return email and delete the original message. Please note, the recipient should check this email and any attachments for the presence of viruses. The organization accepts no liability for any damage caused by any virus transmitted by this email.
=================================

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

mortimus

Re: distance between adjacent particles

In reply to this post by Girault France

This is very relevant to what I am doing at moment, I would like to measure the average the distance of 6 neighboring particles in an array of particles. Does anybody have a plug-in for this?

Thanks.
Mor

Olivier Burri

Re: distance between adjacent particles

Hi all,

While trying out the Script Parameters available, I used this question to make a simple K-Nearest Neighbo(u)rs implementation in Jython

You can find the script here
https://gist.github.com/lacan/2643f2ce7e33d1bb07adafde9ff94101

You need an open image with a multipoint selection to make it work.

The script then asks you for the number of neighbors you want to have and returns a table with the average distance for each point. It also creates an overlay on the image with lines connecting the neighbors and circles whose radii represent the average distance.

Works with calibrated and uncalibrated images.

> -----Original Message-----
> From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
> mortimus
> Sent: mardi, 12 juillet 2016 09:03
> To: [hidden email]
> Subject: Re: distance between adjacent particles
>
> This is very relevant to what I am doing at moment, I would like to measure the
> average the distance of 6 neighboring particles in an array of particles. Does
> anybody have a plug-in for this?
>
> Thanks.
> Mor
>
>
>
> --
> View this message in context: http://imagej.1557.x6.nabble.com/distance-
> between-adjacent-particles-tp3699485p5016864.html
> Sent from the ImageJ mailing list archive at Nabble.com.
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

dobensl

Re: distance between adjacent particles

In reply to this post by Cammer, Michael

this code turns out to be very useful to me - it nicely draws lines between neighboring particles (see below) - I am trying to extend it and measure the Feret’s Angle of that nearest neighbor and, if it is different, wand outline and color that “divergent" nearest neighbor (code below). From my code, when I “wand” the nearest neighbor (Wand(x[shortest1],y[shortest1]);), the value is just the original particle (Wand(x,y)) and no coloring occurs.

Here is the practice image with particles linked by colored lines (very cool to see in action):

[cid:DF89240B-4F5D-409F-9F32-AD1CD5A6BA98]

Here is the code

run("8-bit");
setOption("BlackBackground", true);
run("Make Binary");
run("RGB Color");

/* Next step is to save all 1st and 2nd shortest pairs with their distances.
If distance greater than certain criteria, such as 15% longer than avg dist, then sever it.
Then sort into clusters and do convex hull.

Michael Cammer 20150707 for ImageJ 1.50a
*/

var shortestColor1 = "yellow";
var shortestColor2 = "red";
var shortestColor3 = "blue";
var shortestColor4 = "green";

macro "shortest and 2nd shortest [q]" {
run("Select None");
// This part of the macro gets the XY locations of the points. It could
// be replaced with some other method such as opening a text file or measuring
// centroids of objects by some other method.
run("Clear Results");
run("Set Measurements...", "centroid redirect=None decimal=3");
run("Find Maxima...", "noise=10 output=List");
x = newArray(nResults());
y = newArray(nResults());
for(row=0; row<nResults(); row++) {
x[row] = getResult("X", row);
y[row] = getResult("Y", row);
} // for row to get XY locations

// calculate shortest and 2nd shortest
run("Remove Overlay");
dist = newArray(x.length);
for(i=0; i<x.length; i++) {
for(k=0; k<x.length; k++) {
dist[k] = sqrt(((y[i] - y[k]) * (y[i] - y[k])) + ((x[i] - x[k]) * (x[i] - x[k])));
} // for k inside loop
ranking = Array.rankPositions(dist);
shortest1 = ranking[1]; // ranking[0] will always give the position of 0 in the distance array
shortest2 = ranking[2];
shortest3 = ranking[3];

// uncomment next three lines to show method
// Array.print(dist);
// Array.print(ranking);
// print("");

makeLine(x[i],y[i], x[shortest1],y[shortest1], 1);
Overlay.addSelection(shortestColor1);
makeLine(x[i]+2,y[i]+2, x[shortest2]+2,y[shortest2]+2, 1); // offset to help show when there are shortest & 2nd shortest
Overlay.addSelection(shortestColor2);
makeLine(x[i]-2,y[i]-2, x[shortest3]-2,y[shortest3]-2, 1); // offset to help show when there are shortest & 2nd shortest
Overlay.addSelection(shortestColor3);

/* here are my sad efforts
*/
a1 = newArray(nResults);
a2 = newArray(nResults);

doWand(x[i],y[i]);
roiManager("add");
run("Set Measurements...", "centroid center feret's redirect=None decimal=3");
run("Measure");
wait(200);
a1 = getResult("FeretAngle",i);

doWand(x[shortest1],y[shortest1]);
roiManager("add");
run("Set Measurements...", "centroid center feret's redirect=None decimal=3");
run("Measure");
wait(200);
a2 = getResult("FeretAngle",i);
if (a1 != a2) {

doWand(x[i],y[i]);
run("Add Selection...", "stroke=none width=0 fill=red");
}

}
On May 23, 2016, at 1:05 PM, Cammer, Michael <[hidden email]<mailto:[hidden email]>> wrote:

Curious if anyone thinks this would be useful. It may only need a few lines of code to list the pairs of points and the distances. I wrote it just because I felt like it but no one has brought me a problem to apply it to.
http://microscopynotes.com/imagej/macros/shortest_distances_and_clustering/index.html

=========================================================================
Michael Cammer, Microscopy Core & Skirball Institute, NYU Langone Medical Center
Cell: 914-309-3270 ** Office: Skirball 2nd Floor main office, back right **
http://ocs.med.nyu.edu/microscopy & http://microscopynotes.com/

-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Ben Tupper
Sent: Monday, May 23, 2016 1:08 PM
To: [hidden email]
Subject: Re: distance between adjacent particles

Hi Len,

That's quite an old thread in Nabble. I'm not sure if it will help you, but you could try https://urldefense.proofpoint.com/v2/url?u=http-3A__rsb.info.nih.gov_ij_plugins_graph_index.html&d=CwIFAg&c=j5oPpO0eBH1iio48DtsedbOBGmuw5jHLjgvtN2r4ehE&r=oU_05LztNstAydlbm5L5GDu_vAdjXk3frDLx_CqKkuo&m=UoPrt3a6dSRFDTVKNKhxQxr9hVUnIlg2qHZVtrhy_1M&s=njt95X9CC5z1_RlTbXErpNZ8qshIp4aaCxkviC1GaJI&e= There may be newer tools available that I am not familiar with.

Ben

On May 22, 2016, at 4:06 PM, dobensl <[hidden email]> wrote:

This is a very useful macro and almost takes me to where I want to get
- is there a way to find the values for XM2 and YM2 and add these to
the results table? Also, could someone briefly describe the logic behind the notion:
"initialize it to a large value outside of the for loop"?

Thanks,

Len Dobens

--
View this message in context:
https://urldefense.proofpoint.com/v2/url?u=http-3A__imagej.1557.x6.nab
ble.com_distance-2Dbetween-2Dadjacent-2Dparticles-2Dtp3699485p5016492.
html&d=CwIFAg&c=j5oPpO0eBH1iio48DtsedbOBGmuw5jHLjgvtN2r4ehE&r=oU_05Lzt
NstAydlbm5L5GDu_vAdjXk3frDLx_CqKkuo&m=UoPrt3a6dSRFDTVKNKhxQxr9hVUnIlg2
qHZVtrhy_1M&s=SFCtTsIXuwdtQq58bn7D7GiNgO3u5XcXK5w8iufElds&e=
Sent from the ImageJ mailing list archive at Nabble.com.

--
ImageJ mailing list:
https://urldefense.proofpoint.com/v2/url?u=http-3A__imagej.nih.gov_ij_
list.html&d=CwIFAg&c=j5oPpO0eBH1iio48DtsedbOBGmuw5jHLjgvtN2r4ehE&r=oU_
05LztNstAydlbm5L5GDu_vAdjXk3frDLx_CqKkuo&m=UoPrt3a6dSRFDTVKNKhxQxr9hVU
nIlg2qHZVtrhy_1M&s=A6W4FamOmzV5A4xjupQZMH0LP7G2kO3EEK6U_jH8mbI&e=

--
ImageJ mailing list: https://urldefense.proofpoint.com/v2/url?u=http-3A__imagej.nih.gov_ij_list.html&d=CwIFAg&c=j5oPpO0eBH1iio48DtsedbOBGmuw5jHLjgvtN2r4ehE&r=oU_05LztNstAydlbm5L5GDu_vAdjXk3frDLx_CqKkuo&m=UoPrt3a6dSRFDTVKNKhxQxr9hVUnIlg2qHZVtrhy_1M&s=A6W4FamOmzV5A4xjupQZMH0LP7G2kO3EEK6U_jH8mbI&e=

------------------------------------------------------------
This email message, including any attachments, is for the sole use of the intended recipient(s) and may contain information that is proprietary, confidential, and exempt from disclosure under applicable law. Any unauthorized review, use, disclosure, or distribution is prohibited. If you have received this email in error please notify the sender by return email and delete the original message. Please note, the recipient should check this email and any attachments for the presence of viruses. The organization accepts no liability for any damage caused by any virus transmitted by this email.
=================================

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

Leonard Dobens, PhD
Professor of Molecular Biology and Biochemistry
SCB312
School of Biological Sciences
University of Missouri-Kansas City
5007 Rockhill Road
Kansas City, MO 64110
8162356272
[hidden email]<mailto:[hidden email]>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

!!test6-1.jpeg (57K) Download Attachment