image gradient measurement with ImageJ

[Juan Francisco-2]

Dear all:
  Please i got the following question:
  using ImageJ is possible to create the "gradient" image from a original image?, I mean: for a image given, it would be possible to compute a map showing the gradient on each pixel of the original image? Gradient is the local variation of the signal on the image.
  Thank you very much!!
  Juan Francisco

 
---------------------------------

LLama Gratis a cualquier PC del Mundo.
Llamadas a fijos y móviles desde 1 céntimo por minuto.
http://es.voice.yahoo.com

[Mike Miller-10]

Hello,

My first impression when I think gradient was that you wanted Process ->
Binary -> Distance Map. But since you mentioned the local variance, try
Process -> Filters -> Variance...

This second command will change the color each pixel to correspond with the
variance. You can specify a radius of consideration.

A third consideration might even be able to duplicate the image, perform a
Gaussian blur and then subtract the two images using Process -> Image
Calculator. This last technique has proven very noisy for me, but also
allows more detail through.

I hope one of the last two was helpful.

-Mike

On 3/31/07, Juan Francisco <[hidden email]> wrote:

>
> Dear all:
>   Please i got the following question:
>   using ImageJ is possible to create the "gradient" image from a original
> image?, I mean: for a image given, it would be possible to compute a map
> showing the gradient on each pixel of the original image? Gradient is the
> local variation of the signal on the image.
>   Thank you very much!!
>   Juan Francisco
>
>
> ---------------------------------
>
> LLama Gratis a cualquier PC del Mundo.
> Llamadas a fijos y móviles desde 1 céntimo por minuto.
> http://es.voice.yahoo.com
>

--
-Mike

[Michael Schmid]

Hi Juan,

the gradient (aka Sobel filter) is available as
Process -> Find Edges. It is described at
http://rsb.info.nih.gov/ij/docs/menus/process.html

See also
http://en.wikipedia.org/wiki/Sobel

Michael
________________________________________________________________

On 2 Apr 2007, at 03:39, Mike Miller wrote:

> Hello,
>
> My first impression when I think gradient was that you wanted  
> Process ->
> Binary -> Distance Map. But since you mentioned the local variance,  
> try
> Process -> Filters -> Variance...
>
> This second command will change the color each pixel to correspond  
> with the
> variance. You can specify a radius of consideration.
>
> A third consideration might even be able to duplicate the image,  
> perform a
> Gaussian blur and then subtract the two images using Process -> Image
> Calculator. This last technique has proven very noisy for me, but also
> allows more detail through.
>
> I hope one of the last two was helpful.
>
> -Mike
>
> On 3/31/07, Juan Francisco <[hidden email]> wrote:
>>
>> Dear all:
>>   Please i got the following question:
>>   using ImageJ is possible to create the "gradient" image from a  
>> original
>> image?, I mean: for a image given, it would be possible to compute  
>> a map
>> showing the gradient on each pixel of the original image? Gradient  
>> is the
>> local variation of the signal on the image.
>>   Thank you very much!!
>>   Juan Francisco
>>
>>
>> ---------------------------------
>>
>> LLama Gratis a cualquier PC del Mundo.
>> Llamadas a fijos y móviles desde 1 céntimo por minuto.
>> http://es.voice.yahoo.com
>>
>
>
>
> --
> -Mike

[Hemerson Pistori]

Dear Juan,

     If, besides gradient magnitude, you are interested in
gradient direction information, the following plugin may
help you:

http://bigwww.epfl.ch/thevenaz/differentials/

Best Regards,

Hemerson Pistori
Dom Bosco University
Campo Grande, Brazil
http://www2.acad.ucdb.br/pistori

>> Dear all:
>>  Please i got the following question:
>>  using ImageJ is possible to create the "gradient" image from a original
>> image?, I mean: for a image given, it would be possible to compute a map
>> showing the gradient on each pixel of the original image? Gradient is the
>> local variation of the signal on the image.
>>  Thank you very much!!
>>  Juan Francisco
>>
>>
>> ---------------------------------
>>
>> LLama Gratis a cualquier PC del Mundo.
>> Llamadas a fijos y móviles desde 1 céntimo por minuto.
>> http://es.voice.yahoo.com
>>
>
>
>
> --
> -Mike