Find pattern and mark it with point

[kirill8910]

Hello!

There is a square of a certain pattern in the image.

 

Are there any plugins that could be trained to find the square and mark it with a point right in the middle.

Thank you!

Kirill

[Joel Sheffield]

There is a plugin called "feature finder" at
imagejdocu.tudor.lu/lib/exe/fetch.php?media=plugin:analysis:feature_finder:feature_finder.java

that may do what you want.

Joel Sheffield


On Thu, Mar 15, 2012 at 4:04 AM, kirill8910 <[hidden email]> wrote:

> Hello!
>
> There is a square of a certain pattern in the image.
>
>  http://imagej.1557.n6.nabble.com/file/n4620426/post.png
>
> Are there any plugins that could be trained to find the square and mark it
> with a point right in the middle.
>
> Thank you!
>
> Kirill
>
> --
> View this message in context:
> http://imagej.1557.n6.nabble.com/Find-pattern-and-mark-it-with-point-tp4620426p4620426.html
> Sent from the ImageJ mailing list archive at Nabble.com.
>

--


Joel B. Sheffield, Ph.D
Department of Biology
Temple University
Philadelphia, PA 19122
Voice: 215 204 8839
e-mail: [hidden email]
URL:  http://astro.temple.edu/~jbs

[Joel Sheffield]

My apologies.  Here is the link to the documentation for feature finder.
http://imagejdocu.tudor.lu/doku.php?id=plugin:analysis:feature_finder:start

On Thu, Mar 15, 2012 at 4:04 AM, kirill8910 <[hidden email]> wrote:

> Hello!
>
> There is a square of a certain pattern in the image.
>
>  http://imagej.1557.n6.nabble.com/file/n4620426/post.png
>
> Are there any plugins that could be trained to find the square and mark it
> with a point right in the middle.
>
> Thank you!
>
> Kirill
>
> --
> View this message in context:
> http://imagej.1557.n6.nabble.com/Find-pattern-and-mark-it-with-point-tp4620426p4620426.html
> Sent from the ImageJ mailing list archive at Nabble.com.
>

--


Joel B. Sheffield, Ph.D
Department of Biology
Temple University
Philadelphia, PA 19122
Voice: 215 204 8839
e-mail: [hidden email]
URL:  http://astro.temple.edu/~jbs

[kirill8910]

Yes, super!

Thanks a lot, Joel!

Kirill

[Cammer, Michael]

I don't know about "training", but if there are certain rules or boundaries such as the object of interest (for instance):
is always be contained entirely within the image;
is significantly brighter and/or darker than the material around it;
is less saturated than the material around it;
has certain shape descriptors,
then ImageJ (macro or plugin) can very easily be told to follow those rules to find similar objects in similar pictures.

In this example the components of the square are either lighter or darker than everything else.  Also, they are less saturated than everything else.  Also, the components of the big square are themselves square.  And the components are close together.  So it is easy for the software to identify the square (and toss out other features that may be selected by one or two of  the three criteria but not all three).  The challenge is the other n pictures that need to have the features automatically detected based on these rules.  

Can ImageJ be taught to differentiate this image from Sean Scully paintings?

________________________________________________________
Michael Cammer, Assistant Research Scientist
Skirball Institute of Biomolecular Medicine
Lab: [phone]  Cell: [phone]


-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of kirill8910
Sent: Thursday, March 15, 2012 4:05 AM
To: [hidden email]
Subject: Find pattern and mark it with point

Hello!

There is a square of a certain pattern in the image.

  http://imagej.1557.n6.nabble.com/file/n4620426/post.png 

Are there any plugins that could be trained to find the square and mark it
with a point right in the middle.

Thank you!

Kirill

--
View this message in context: http://imagej.1557.n6.nabble.com/Find-pattern-and-mark-it-with-point-tp4620426p4620426.html
Sent from the ImageJ mailing list archive at Nabble.com.

[dscho]

Hi Kirill,

On Thu, 15 Mar 2012, kirill8910 wrote:

> There is a square of a certain pattern in the image.
>
>   http://imagej.1557.n6.nabble.com/file/n4620426/post.png 
>
> Are there any plugins that could be trained to find the square and mark
> it with a point right in the middle.

You might want to look at Visual grep, included in the Fiji distribution
of ImageJ. It requires an image of what you want to find (needle) in
addition to the image in which you want to find it. It will not mark
things as dots but as selections (if it finds multiple candidates -- which
will be more likely when you increase the tolerance -- the selection will
be a composite of rectangles).

Ciao,
Johannes

[kirill8910]

Interesting stuff, Michael!

But I am not sure If I understand abot Sean Scully paintings...they also look like to have patterns that could be recognized..

Kirill

15.03.2012, 16:28, "Cammer, Michael" <[hidden email]>:

> I don't know about "training", but if there are certain rules or boundaries such as the object of interest (for instance):
> is always be contained entirely within the image;
> is significantly brighter and/or darker than the material around it;
> is less saturated than the material around it;
> has certain shape descriptors,
> then ImageJ (macro or plugin) can very easily be told to follow those rules to find similar objects in similar pictures.
>
> In this example the components of the square are either lighter or darker than everything else.  Also, they are less saturated than everything else.  Also, the components of the big square are themselves square.  And the components are close together.  So it is easy for the software to identify the square (and toss out other features that may be selected by one or two of  the three criteria but not all three).  The challenge is the other n pictures that need to have the features automatically detected based on these rules.
>
> Can ImageJ be taught to differentiate this image from Sean Scully paintings?
>
> ________________________________________________________
> Michael Cammer, Assistant Research Scientist
> Skirball Institute of Biomolecular Medicine
> Lab: [phone]  Cell: [phone]
>
> -----Original Message-----
> From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of kirill8910
> Sent: Thursday, March 15, 2012 4:05 AM
> To: [hidden email]
> Subject: Find pattern and mark it with point
>
> Hello!
>
> There is a square of a certain pattern in the image.
>
>   http://imagej.1557.n6.nabble.com/file/n4620426/post.png
>
> Are there any plugins that could be trained to find the square and mark it
> with a point right in the middle.
>
> Thank you!
>
> Kirill
>
> --
> View this message in context: http://imagej.1557.n6.nabble.com/Find-pattern-and-mark-it-with-point-tp4620426p4620426.html
> Sent from the ImageJ mailing list archive at Nabble.com.

[lechristophe]

Hi,

There is also the Template Matching plugin developped by Qingzong TSENG :
https://sites.google.com/site/qingzongtseng/template-matching-ij-plugin

Cheers,

Christophe



2012/3/16 Mst Kirill <[hidden email]>:

> Interesting stuff, Michael!
>
> But I am not sure If I understand abot Sean Scully paintings...they also look like to have patterns that could be recognized..
>
> Kirill
>
> 15.03.2012, 16:28, "Cammer, Michael" <[hidden email]>:
>> I don't know about "training", but if there are certain rules or boundaries such as the object of interest (for instance):
>> is always be contained entirely within the image;
>> is significantly brighter and/or darker than the material around it;
>> is less saturated than the material around it;
>> has certain shape descriptors,
>> then ImageJ (macro or plugin) can very easily be told to follow those rules to find similar objects in similar pictures.
>>
>> In this example the components of the square are either lighter or darker than everything else.  Also, they are less saturated than everything else.  Also, the components of the big square are themselves square.  And the components are close together.  So it is easy for the software to identify the square (and toss out other features that may be selected by one or two of  the three criteria but not all three).  The challenge is the other n pictures that need to have the features automatically detected based on these rules.
>>
>> Can ImageJ be taught to differentiate this image from Sean Scully paintings?
>>
>> ________________________________________________________
>> Michael Cammer, Assistant Research Scientist
>> Skirball Institute of Biomolecular Medicine
>> Lab: [phone]  Cell: [phone]
>>
>> -----Original Message-----
>> From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of kirill8910
>> Sent: Thursday, March 15, 2012 4:05 AM
>> To: [hidden email]
>> Subject: Find pattern and mark it with point
>>
>> Hello!
>>
>> There is a square of a certain pattern in the image.
>>
>>   http://imagej.1557.n6.nabble.com/file/n4620426/post.png
>>
>> Are there any plugins that could be trained to find the square and mark it
>> with a point right in the middle.
>>
>> Thank you!
>>
>> Kirill
>>
>> --
>> View this message in context: http://imagej.1557.n6.nabble.com/Find-pattern-and-mark-it-with-point-tp4620426p4620426.html
>> Sent from the ImageJ mailing list archive at Nabble.com.

[kirill8910]

Thanks Christophe!

16.03.2012, 14:20, "Christophe Leterrier" <[hidden email]>:

> Hi,
>
> There is also the Template Matching plugin developped by Qingzong TSENG :
> https://sites.google.com/site/qingzongtseng/template-matching-ij-plugin
>
> Cheers,
>
> Christophe
>
> 2012/3/16 Mst Kirill <[hidden email]>:
>
>>  Interesting stuff, Michael!
>>
>>  But I am not sure If I understand abot Sean Scully paintings...they also look like to have patterns that could be recognized..
>>
>>  Kirill
>>
>>  15.03.2012, 16:28, "Cammer, Michael" <[hidden email]>:
>>>  I don't know about "training", but if there are certain rules or boundaries such as the object of interest (for instance):
>>>  is always be contained entirely within the image;
>>>  is significantly brighter and/or darker than the material around it;
>>>  is less saturated than the material around it;
>>>  has certain shape descriptors,
>>>  then ImageJ (macro or plugin) can very easily be told to follow those rules to find similar objects in similar pictures.
>>>
>>>  In this example the components of the square are either lighter or darker than everything else.  Also, they are less saturated than everything else.  Also, the components of the big square are themselves square.  And the components are close together.  So it is easy for the software to identify the square (and toss out other features that may be selected by one or two of  the three criteria but not all three).  The challenge is the other n pictures that need to have the features automatically detected based on these rules.
>>>
>>>  Can ImageJ be taught to differentiate this image from Sean Scully paintings?
>>>
>>>  ________________________________________________________
>>>  Michael Cammer, Assistant Research Scientist
>>>  Skirball Institute of Biomolecular Medicine
>>>  Lab: [phone]  Cell: [phone]
>>>
>>>  -----Original Message-----
>>>  From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of kirill8910
>>>  Sent: Thursday, March 15, 2012 4:05 AM
>>>  To: [hidden email]
>>>  Subject: Find pattern and mark it with point
>>>
>>>  Hello!
>>>
>>>  There is a square of a certain pattern in the image.
>>>
>>>    http://imagej.1557.n6.nabble.com/file/n4620426/post.png
>>>
>>>  Are there any plugins that could be trained to find the square and mark it
>>>  with a point right in the middle.
>>>
>>>  Thank you!
>>>
>>>  Kirill
>>>
>>>  --
>>>  View this message in context: http://imagej.1557.n6.nabble.com/Find-pattern-and-mark-it-with-point-tp4620426p4620426.html
>>>  Sent from the ImageJ mailing list archive at Nabble.com.