imlab-users — IMLAB Users Discussion List

[Imlab-users] ImLab 3.1

[Antonio S. <ant...@gm...>]

  Hi,

  ImLab 3.1 was released.

  In this version we fixed a few performance problems, along with other
minor problems. And we would like to highlight the following changes:

* There is a new Set Scale dialog in Analysis, and scale support in
results. A long waited feature.
* The Histogram Window now support for SHORT and USHORT.
* New option for result using double precision in Process / Color
Components / Split HSI and Normalize Components, Process / Remote Sensing /
Normalized Difference Ratio, Process / Domain Transform / Distance
Transform and Forward Fourier and Cross Correlation and Auto Correlation.
* Improved precision in several operations by using double numeric
precision.
* Changed to FFTW version 3 and adds double support for FFT transform
operations.

Documentation can be found at:
http://imlab.sourceforge.net/

Best Regards,
Antonio Scuri

[Imlab-users] ImLab 3.0

[Antonio S. <ant...@gm...>]

Hi,

ImLab 3.0 is released.

This is major update with a new interface, but also updates to the new
versions of IUP, CD and IM libraries.

ImLab can be found at:
http://imlab.sourceforge.net/

Best Regards,
Antonio Scuri

Re: [Imlab-users] frequency spectrum of an image

[Antonio S. <sc...@te...>]

  Hi,

 

  Images are 2D functions by definition, I mean each gray or color value is
f(x,y).

 

  So its frequency distribution is also a 2D function, or also an image. In
other words, images have spatial frequencies, so they cannot be represented
in a simple graph as you want.

 

  That's why the menu "Process/Domain Transfom/Forward Fourier" produces
also an image.

 

Best,

Scuri

 

 

From: Shah Khalid [mailto:cs....@gm...] 
Sent: quarta-feira, 3 de agosto de 2011 06:09
To: Antonio Scuri
Cc: iml...@li...
Subject: Re: [Imlab-users] frequency spectrum of an image

 

Dear Antonio,

Thank you very much for the suggestion.

Actually I want to see a plot where the frequency(Hz) of the image is on x
axis and its amplitude or sepectral power on y axis. Is it possible?

Best regards,
Shah.

On Tue, Aug 2, 2011 at 8:48 PM, Antonio Scuri <sc...@te...>
wrote:

 Hi Shah,

 

  To inspect the frequency distribution of an image, the best tool I know is
the Fourier Transform.

 

  You can use the "Process/Domain Transfom/Forward Fourier" menu to see the
frequency distribution.

 

Best Regards,

Antonio Scuri

 

From: Shah Khalid [mailto:cs....@gm...] 
Sent: segunda-feira, 1 de agosto de 2011 09:36
To: iml...@li...
Subject: [Imlab-users] frequency spectrum of an image

 

Dear all,

I use high-pass (LoG) and low-pass (Guassian) filtering function of imlab.
Thanks for the nice software.

I need to draw the frequency distribution plots of the images before and
after filtering, to show
which frequency bands were present before and after filtering.

Can you please help me in this issue?

I played around the "Find and show histogram view", but I'm not quite sure
it is the right function
as the ouput on x and y coordinates are not labeled. And I doubt that it is
simply the histogram count
of the number of pixels having specific intensity?

Thank you very much in advance for your help and quick reply.

Best regards,
Shah.

 

Re: [Imlab-users] frequency spectrum of an image

[Shah K. <cs....@gm...>]

Dear Antonio,

Thank you very much for the suggestion.

Actually I want to see a plot where the frequency(Hz) of the image is on x
axis and its amplitude or sepectral power on y axis. Is it possible?

Best regards,
Shah.

On Tue, Aug 2, 2011 at 8:48 PM, Antonio Scuri <sc...@te...>wrote:

>  Hi Shah,****
>
> ** **
>
>   To inspect the frequency distribution of an image, the best tool I know
> is the Fourier Transform.****
>
> ** **
>
>   You can use the “Process/Domain Transfom/Forward Fourier” menu to see the
> frequency distribution.****
>
> ** **
>
> Best Regards,****
>
> Antonio Scuri****
>
> ** **
>
> *From:* Shah Khalid [mailto:cs....@gm...]
> *Sent:* segunda-feira, 1 de agosto de 2011 09:36
> *To:* iml...@li...
> *Subject:* [Imlab-users] frequency spectrum of an image****
>
> ** **
>
> Dear all,
>
> I use high-pass (LoG) and low-pass (Guassian) filtering function of imlab.
> Thanks for the nice software.
>
> I need to draw the frequency distribution plots of the images before and
> after filtering, to show
> which frequency bands were present before and after filtering.
>
> Can you please help me in this issue?
>
> I played around the "Find and show histogram view", but I'm not quite sure
> it is the right function
> as the ouput on x and y coordinates are not labeled. And I doubt that it is
> simply the histogram count
> of the number of pixels having specific intensity?
>
> Thank you very much in advance for your help and quick reply.
>
> Best regards,
> Shah.****
>

Re: [Imlab-users] frequency spectrum of an image

[Antonio S. <sc...@te...>]

 Hi Shah,

 

  To inspect the frequency distribution of an image, the best tool I know is
the Fourier Transform.

 

  You can use the "Process/Domain Transfom/Forward Fourier" menu to see the
frequency distribution.

 

Best Regards,

Antonio Scuri

 

From: Shah Khalid [mailto:cs....@gm...] 
Sent: segunda-feira, 1 de agosto de 2011 09:36
To: iml...@li...
Subject: [Imlab-users] frequency spectrum of an image

 

Dear all,

I use high-pass (LoG) and low-pass (Guassian) filtering function of imlab.
Thanks for the nice software.

I need to draw the frequency distribution plots of the images before and
after filtering, to show
which frequency bands were present before and after filtering.

Can you please help me in this issue?

I played around the "Find and show histogram view", but I'm not quite sure
it is the right function
as the ouput on x and y coordinates are not labeled. And I doubt that it is
simply the histogram count
of the number of pixels having specific intensity?

Thank you very much in advance for your help and quick reply.

Best regards,
Shah.

[Imlab-users] frequency spectrum of an image

[Shah K. <cs....@gm...>]

Dear all,

I use high-pass (LoG) and low-pass (Guassian) filtering function of imlab.
Thanks for the nice software.

I need to draw the frequency distribution plots of the images before and
after filtering, to show
which frequency bands were present before and after filtering.

Can you please help me in this issue?

I played around the "Find and show histogram view", but I'm not quite sure
it is the right function
as the ouput on x and y coordinates are not labeled. And I doubt that it is
simply the histogram count
of the number of pixels having specific intensity?

Thank you very much in advance for your help and quick reply.

Best regards,
Shah.

[Imlab-users] IMLAB 2.3.4

[Antonio S. <sc...@te...>]

Hi,

ImLab 2.3.4 is released.

This is again a small update with a few new features, but also updates to
the new versions of IUP, CD and IM libraries.

ImLab can be found at:

http://imlab.sourceforge.net/

ImLab is a free open source graphical application for Scientific Image
Processing that runs in Windows, Linux and many other UNIX systems. It
supports multiple windows, data types including 32 bit integers,
32 bit real numbers and complex numbers. It is implemented in C++ and also
in C to provide a very simple way to add new functions. It has many image
operations and supports several file formats.

Best Regards,
Antonio Scuri

[Imlab-users] ImLab 2.3.3

[Antonio S. <sc...@te...>]

  Hi,

  ImLab 2.3.3 is released.

  This is a small update with a few new features, but also updates to the
new versions of IUP and IM libraries.

ImLab can be found at:

     http://imlab.sourceforge.net/

  ImLab is a free open source graphical application for Scientific Image
Processing that runs in Windows, Linux and many other UNIX systems. It
supports multiple windows, data types including 32 bit integers,
32 bit real numbers and complex numbers. It is implemented in C++ and also
in C to provide a very simple way to add new functions. It has many image
operations and supports several file formats.

Best Regards,
Antonio Scuri

[Imlab-users] IMLAB 2.3.2

[Antonio S. <sc...@te...>]

  Hi,

  IMLAB 2.3.2 is released.

  This is a small update to use the
new versions of IUP, CD and IM libraries.

ImLab can be found at:

     http://imlab.sourceforge.net/

  If IMLAB is new to you it is a free open source graphical application for
Scientific Image Processing that runs in Windows, Linux and many other UNIX
systems. It supports multiple windows, data types including 32 bit integers,
32 bit real numbers and complex numbers. It is implemented in C++ and also
in C to provide a very simple way to add new functions. It has many image
operations and supports several file formats.

Best Regards,
Antonio Scuri

Re: [Imlab-users] Information on ImLab

[Antonio S. <sc...@te...>]

  Hi Pallavi,

> Respected Sir,
> 
> I'm Pallavi, an Engineering passing-out student keen on learning machine
> vision application and development.
> During my search for a machine vision free software equivalent to NI
Vision
> Builder and Development toolkit, I came across your work on ImLab. The
> intention of this mail is to get some clarifications about the software.
Here
> are my queries :
> 
> -- Is it really a freeware?

  Yes.


> -- Is ImLab meant only for face recognition or can it be used for any sort
of
> image recognition?

  It can be used for any sort of image recognition. Just curious, where did
you had the impression that it is meant only for face recognition?


> -- Can the software be used for any sort of image training and image
reading?

  Don't know what you mean by "image training".


> -- Can it perform OCR i.e Optical Character Recognition?

  No. You will have to use an external toolkit along with IMLAB to do that.
It is possible.
 

> Ultimately my intention is to know whether can it be used as a replacement
> for NI Vision Builder and Development toolkit?

  That's something difficult to answer. In a general way the answer is not,
although I don't know much about NI Vision Builder, as been a commercial
toolkit it usually has more investment than IMLAB. Partially IMLAB can
replace it.

 
> If not please let me know if there is any free i.e an open-source
equivalent
> to NI Vision Builder and Development toolkit?

  No I don't.

 
> It would be very kind and helpful of you if you could spare a bit of your
> valuable and precious time in responding me with the required information.
> 
> Anticipating your reply as early as possible.
> 
> Thanks and Regards,
> Pallavi.
> +91 9741911633.
> Bangalore.
> INDIA.

Best Regards,
Antonio Scuri

[Imlab-users] Information on ImLab

[pallavi k. <pal...@gm...>]

Respected Sir,

I'm Pallavi, an Engineering passing-out student keen on learning machine
vision application and development.
During my search for a machine vision free software equivalent to NI Vision
Builder and Development toolkit,
I came across your work on *ImLab*. The intention of this mail is to get
some clarifications about the
software. Here are my queries :

-- Is it really a freeware?
-- Is *ImLab* meant only for face recognition or can it be used for any sort
of image recognition?
-- Can the software be used for any sort of image training and image
reading?
-- Can it perform OCR i.e Optical Character Recognition?

Ultimately my intention is to know whether can it be used as a replacement
for NI Vision Builder and Development toolkit?

If not please let me know if there is any free i.e *an open-source
equivalent to NI Vision Builder and Development toolkit*?

It would be very kind and helpful of you if you could spare a bit of your
valuable and precious time in responding me with
the required information.

Anticipating your reply as early as possible.

Thanks and Regards,
Pallavi.
+91 9741911633.
Bangalore.
INDIA.

[Imlab-users] IMLAB 2.3.1

[Antonio S. <sc...@te...>]

  Hi,

  IMLAB 2.3.1 is released.

  This is a small update with a few new features, but also updates to the
new versions of IUP, CD and IM libraries.

ImLab can be found at:

     http://imlab.sourceforge.net/

  If IMLAB is new to you it is a free open source graphical application for
Scientific Image Processing that runs in Windows, Linux and many other UNIX
systems. It supports multiple windows, data types including 32 bit integers,
32 bit real numbers and complex numbers. It is implemented in C++ and also
in C to provide a very simple way to add new functions. It has many image
operations and supports several file formats.

Best Regards,
Antonio Scuri

Re: [Imlab-users] Incoplete frame

[<bio...@gm...>]

Hi all,

i try the Scuri tip to do the program Sleep...
i put  100 miliseconds to sleep, the image come perfect  switching for 
evrey 4 inputs...
but this made just 10 frames per sec....i need 30..
this can be a limitation of hardware ?
there another thing i can do ?



Antonio Scuri escreveu:
>   Hi,
>
>   The imVideoCapture capture implementation should guarantee that every time
> you call imVideoCaptureFrame you got a new frame or it will return 0.
>
>   If all returns are 1, then usually could be two problems:
>
>   - you are capturing during device initialization, so there can be trash in
> the device buffer.
>
> Or 
>
>   - you are capturing during input switch.
>
>   In both cases, I can recommend only to use a delay to give some time to
> the hardware respond.
>
>   Direct X allows that situations to happen. We have to figure out what to
> do...
>
> Best Regards,
> Antonio Scuri
>
>   
>> -----Original Message-----
>> From: iml...@li... 
>> [mailto:iml...@li...] On Behalf 
>> Of bio...@gm...
>> Sent: segunda-feira, 19 de novembro de 2007 17:34
>> To: iml...@li...
>> Subject: [Imlab-users] Incoplete frame
>>
>> Hi all,
>>
>> I using IM lab to capture frame from a device with 4 inputs, 
>> i using the imVideoCaptureSetInOut to catch a frame for all 
>> inputs.... 
>> and using imVideoCaptureFrame to catch the frames....
>> but...the frames come fuzzy,
>> how i know the camera give to me a complete frame(or a new) ?
>>
>> thanks..
>>
>> --------------------------------------------------------------
>> -----------
>> This SF.net email is sponsored by: Microsoft Defy all 
>> challenges. Microsoft(R) Visual Studio 2005.
>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>> _______________________________________________
>> Imlab-users mailing list
>> Iml...@li...
>> https://lists.sourceforge.net/lists/listinfo/imlab-users
>>
>>     
>
>
>   

Re: [Imlab-users] Incoplete frame

[<bio...@gm...>]

i will try this, thanks...
by the way...i used your Fundamentos da Imagem Digital in my final work 
for graduation, is very good!


Antonio Scuri escreveu:
>   Hi,
>
>   The imVideoCapture capture implementation should guarantee that every time
> you call imVideoCaptureFrame you got a new frame or it will return 0.
>
>   If all returns are 1, then usually could be two problems:
>
>   - you are capturing during device initialization, so there can be trash in
> the device buffer.
>
> Or 
>
>   - you are capturing during input switch.
>
>   In both cases, I can recommend only to use a delay to give some time to
> the hardware respond.
>
>   Direct X allows that situations to happen. We have to figure out what to
> do...
>
> Best Regards,
> Antonio Scuri
>
>   
>> -----Original Message-----
>> From: iml...@li... 
>> [mailto:iml...@li...] On Behalf 
>> Of bio...@gm...
>> Sent: segunda-feira, 19 de novembro de 2007 17:34
>> To: iml...@li...
>> Subject: [Imlab-users] Incoplete frame
>>
>> Hi all,
>>
>> I using IM lab to capture frame from a device with 4 inputs, 
>> i using the imVideoCaptureSetInOut to catch a frame for all 
>> inputs.... 
>> and using imVideoCaptureFrame to catch the frames....
>> but...the frames come fuzzy,
>> how i know the camera give to me a complete frame(or a new) ?
>>
>> thanks..
>>
>> --------------------------------------------------------------
>> -----------
>> This SF.net email is sponsored by: Microsoft Defy all 
>> challenges. Microsoft(R) Visual Studio 2005.
>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>> _______________________________________________
>> Imlab-users mailing list
>> Iml...@li...
>> https://lists.sourceforge.net/lists/listinfo/imlab-users
>>
>>     
>
>
>   

Re: [Imlab-users] Incoplete frame

[Antonio S. <sc...@te...>]

  Hi,

  The imVideoCapture capture implementation should guarantee that every time
you call imVideoCaptureFrame you got a new frame or it will return 0.

  If all returns are 1, then usually could be two problems:

  - you are capturing during device initialization, so there can be trash in
the device buffer.

Or 

  - you are capturing during input switch.

  In both cases, I can recommend only to use a delay to give some time to
the hardware respond.

  Direct X allows that situations to happen. We have to figure out what to
do...

Best Regards,
Antonio Scuri

> -----Original Message-----
> From: iml...@li... 
> [mailto:iml...@li...] On Behalf 
> Of bio...@gm...
> Sent: segunda-feira, 19 de novembro de 2007 17:34
> To: iml...@li...
> Subject: [Imlab-users] Incoplete frame
> 
> Hi all,
> 
> I using IM lab to capture frame from a device with 4 inputs, 
> i using the imVideoCaptureSetInOut to catch a frame for all 
> inputs.... 
> and using imVideoCaptureFrame to catch the frames....
> but...the frames come fuzzy,
> how i know the camera give to me a complete frame(or a new) ?
> 
> thanks..
> 
> --------------------------------------------------------------
> -----------
> This SF.net email is sponsored by: Microsoft Defy all 
> challenges. Microsoft(R) Visual Studio 2005.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Imlab-users mailing list
> Iml...@li...
> https://lists.sourceforge.net/lists/listinfo/imlab-users
> 

[Imlab-users] Incoplete frame

[<bio...@gm...>]

Hi all,

I using IM lab to capture frame from a device with 4 inputs,
i using the imVideoCaptureSetInOut to catch a frame for all inputs.... 
and using imVideoCaptureFrame to catch the frames....
but...the frames come fuzzy,
how i know the camera give to me a complete frame(or a new) ?

thanks..

Re: [Imlab-users] Fwd: IM 3.1 compiling on ARM

[Antonio S. <sc...@te...>]

At 15:03 19/7/2006, Jesse Schwartzentruber wrote:
>I am trying to use IM on an ARM platform, and I can make it compile
>okay, except for warnings such as the following:
>im_converttype.cpp
>im_converttype.cpp: In function `int iConvertInt2Int(int, const SRCT*,
>DSTT*, int, int, int) [with SRCT = imbyte, DSTT = imushort]':
>im_converttype.cpp:392:   instantiated from here
>im_converttype.cpp:130: warning: comparison is always true due to
>limited range of data type
>im_converttype.cpp:130: warning: comparison is always true due to
>limited range of data type
>im_converttype.cpp: In function `int iConvertInt2Int(int, const SRCT*,
>DSTT*, int, int, int) [with SRCT = imbyte, DSTT = int]':
>im_converttype.cpp:398:   instantiated from here
>im_converttype.cpp:130: warning: comparison is always true due to
>limited range of data type
>im_converttype.cpp:130: warning: comparison is always true due to
>limited range of data type
>I know C very well, but I don't understand the C++ code here. I don't
>know why these comparisons are failing.

   That warnings should be harmless. Do not worry about them.


>The result of using the library is that I can do some things, such 
>as loading files and saving
>them in different formats, but when I try to change an RGB image to
>BINARY or GRAY, for example, the image comes out almost completely
>black, but using imlab on the desktop to convert the same image to
>binary works fine.

   Notice that the warnings are on "im_converttype.cpp" that do not 
affect the color convertions in "im_convertcolor.cpp". Any warnings 
on that module?

Best,
scuri



    

[Imlab-users] Fwd: IM 3.1 compiling on ARM

[Jesse S. <jd...@gm...>]

I am trying to use IM on an ARM platform, and I can make it compile
okay, except for warnings such as the following:

im_converttype.cpp
im_converttype.cpp: In function `int iConvertInt2Int(int, const SRCT*,
DSTT*, int, int, int) [with SRCT = imbyte, DSTT = imushort]':
im_converttype.cpp:392:   instantiated from here
im_converttype.cpp:130: warning: comparison is always true due to
limited range of data type
im_converttype.cpp:130: warning: comparison is always true due to
limited range of data type
im_converttype.cpp: In function `int iConvertInt2Int(int, const SRCT*,
DSTT*, int, int, int) [with SRCT = imbyte, DSTT = int]':
im_converttype.cpp:398:   instantiated from here
im_converttype.cpp:130: warning: comparison is always true due to
limited range of data type
im_converttype.cpp:130: warning: comparison is always true due to
limited range of data type

I know C very well, but I don't understand the C++ code here. I don't
know why these comparisons are failing. The result of using the
library is that I can do some things, such as loading files and saving
them in different formats, but when I try to change an RGB image to
BINARY or GRAY, for example, the image comes out almost completely
black, but using imlab on the desktop to convert the same image to
binary works fine.

-Jesse

[Imlab-users] IMLAB 2.3

[Antonio S. <sc...@te...>]

   Hi,

   IMLAB 2.3 is released.

   Now in a new web site at SourceForge, with CVS, mailing list and 
mirrored downloads. The new page is at:

      http://imlab.sourceforge.net/

   The old page will automatically redirect you to the new one.

   There are a few changes in IMLAB itself, but one new very 
interesting, a 3D view of the image plane using intensity as the 
third dimension.

   Thanks for everyone for the feedback on IMLAB.

Best Regards,
Antonio Scuri