I have the similar problem but it is at stage 7.
Anyone can help?
--- subsins <subsins@...> wrote:

> hye all
> i think beta 4 got some problems with in cause i get
> memory could not
> be written at stage 0 can anyone point to this
> problem what is the
> problem with haartraining.
> thanks in advance,
> khuram.
>
>
>
> ------------------------ Yahoo! Groups Sponsor
> --------------------~-->
> Make a clean sweep of pop-up ads. Yahoo! Companion
> Toolbar.
> Now with Pop-Up Blocker. Get it for free!
>
http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/W4wwlB/TM
>
--------------------------------------------------------------------~->
>
>
> Change settings:
> http://www.yahoogroups.com/mygroups, select
>    Get Emails (get all posts)
>    Daily Digest (one summary email per day)
>    Read on the web (read posts on the web only)Or
> Unsubscribe by mailing
> OpenCV-unsubscribe@yahoogroups.com
>
> Yahoo! Groups Links
>
>
>     OpenCV-unsubscribe@yahoogroups.com
>
>
>
>




__________________________________
Do you Yahoo!?
Y! Messenger - Communicate in real time. Download now.
http://messenger.yahoo.com

Hi
I'm trying to call OpenCV from Octave (a matlab like enviroment) but I'm
having problems with the OpenCV error handler. As an example
I want to call cvDilate with a image of 64 bit depth. This will raise an
error since cvDilate doesn't support 64 bit images. I handle this with
the following code:

/* Set the error mode */
cvSetErrMode( CV_ErrModeSilent );

/* Perform the actual function call */
cvDilate( input0, output0, input1, input2);

/* Check for errors */
if ( cvGetErrStatus() < 0 ) {
	 error("OpenCV returned the following error:");
	 error( cvErrorStr(cvGetErrStatus()) );

	 /* Deallocate the created variables */
	 cvReleaseImage( &input0 );
	 cvReleaseStructuringElement( &input1 );
	 cvReleaseImage( &output0 );

	 return out_list;
}

And this kinda works. The first time I call the function I get the error
message, but the next time I get a segfault. I assume this is because I
don't handle the error in a proper manor.

Can anybody help me here?

Søren

Thank you very much!
The problem is that I have windows xp.
Thanks anyway!
Juan

--- In OpenCV@yahoogroups.com, "tkwoodfamily" <yahoo@k...> wrote:
> Linux or Windows?
>
> Linux Beta4? Go to file:
> /opencv/otherlibs/highgui/cvcap_v4l.cpp
> take out the double slash (//) that will un comment the line:
> #define V4L_SCALE_OPTION 1
> also change the DEFAULT_WIDTH and DEFAULT_HEIGHT to your prefered
size.
> then -
>
> make
> make install
>
> Re run and enjoy
>
>
> Windows Beta4?  Dont know.  Someone else will have to help you.
>
> Travis Wood
>
>
>
>
> --- In OpenCV@yahoogroups.com, "juanwachs" <juan@b...> wrote:
> > Dear OpenCV users:
> > I'm trying to modify the following code to increase the window of
the
> > web-cam to 640x480. I know that there is an application
> > called 'cvcamdemo' inside the cvcam folder of the OpenCV, which
> > change the window size through a program, but I have problems
adding
> > new windows there.
> > So I prefer to change the following code. Why the commands
> >    cvSetCaptureProperty(capture,CV_CAP_PROP_FRAME_WIDTH , 640)
and
> >     cvSetCaptureProperty(capture,CV_CAP_PROP_FRAME_HEIGHT , 480);
> > are not working???
> > Need help!
> >
> > int main(int argc, char* argv[])
> > {
> >  int out=0;
> >  IplImage* laplace = 0;
> >     IplImage* colorlaplace = 0;
> >     IplImage* planes[3] = { 0, 0, 0 };
> >     CvCapture* capture = 0;
> >
> >
> >     if( argc == 1 || (argc == 2 && strlen(argv[1]) == 1 && isdigit
> > (argv[1][0])))
> >         capture = cvCaptureFromCAM( argc == 2 ? argv[1][0] -
  '0' :
> > 0 );
> >     else if( argc == 2 )
> >         capture = cvCaptureFromAVI( argv[1] );
> >
> >     if( !capture )
> >     {
> >         fprintf(stderr,"Could not initialize capturing...\n");
> >         return -1;
> >     }
> >
> >     cvNamedWindow( "R", CV_WINDOW_AUTOSIZE );
> >     cvSetCaptureProperty(capture,CV_CAP_PROP_FRAME_WIDTH , 640);
> >     cvSetCaptureProperty(capture,CV_CAP_PROP_FRAME_HEIGHT , 480);
> >
> >
> >
> >     for(;;)
> >     {
> >         IplImage* frame = 0;
> >         int i;
> >
> >         frame = cvQueryFrame( capture );
> >         if( !frame )
> >             break;
> >
> >             for( i = 0; i < 3; i++ )
> >                 planes[i] = cvCreateImage( cvSize(frame-
>width,frame-
> > >height), 8, 1 );
> >
> >         cvCvtPixToPlane( frame, planes[0], planes[1], planes[2],
0 );
> >
> > 		    for( i = 0; i < 3; i++ )
> > 		       planes[i]->origin = frame->origin;
> >
> >         cvShowImage("R", planes[0]);
> >
> >         if( cvWaitKey(10) >= 0 )
> >             break;
> >     }
> >
> >     cvReleaseCapture( &capture );
> >     cvDestroyWindow("Video-Window");
> >
> >     return 0;
> > }

Well! It worked also for the windows xp.
I follow your instructions and it works!
Thanks!
Juan


-- In OpenCV@yahoogroups.com, "tkwoodfamily" <yahoo@k...> wrote:
> Linux or Windows?
>
> Linux Beta4? Go to file:
> /opencv/otherlibs/highgui/cvcap_v4l.cpp
> take out the double slash (//) that will un comment the line:
> #define V4L_SCALE_OPTION 1
> also change the DEFAULT_WIDTH and DEFAULT_HEIGHT to your prefered
size.
> then -
>
> make
> make install
>
> Re run and enjoy
>
>
> Windows Beta4?  Dont know.  Someone else will have to help you.
>
> Travis Wood
>
>
>
>
> --- In OpenCV@yahoogroups.com, "juanwachs" <juan@b...> wrote:
> > Dear OpenCV users:
> > I'm trying to modify the following code to increase the window of
the
> > web-cam to 640x480. I know that there is an application
> > called 'cvcamdemo' inside the cvcam folder of the OpenCV, which
> > change the window size through a program, but I have problems
adding
> > new windows there.
> > So I prefer to change the following code. Why the commands
> >    cvSetCaptureProperty(capture,CV_CAP_PROP_FRAME_WIDTH , 640)
and
> >     cvSetCaptureProperty(capture,CV_CAP_PROP_FRAME_HEIGHT , 480);
> > are not working???
> > Need help!
> >
> > int main(int argc, char* argv[])
> > {
> >  int out=0;
> >  IplImage* laplace = 0;
> >     IplImage* colorlaplace = 0;
> >     IplImage* planes[3] = { 0, 0, 0 };
> >     CvCapture* capture = 0;
> >
> >
> >     if( argc == 1 || (argc == 2 && strlen(argv[1]) == 1 && isdigit
> > (argv[1][0])))
> >         capture = cvCaptureFromCAM( argc == 2 ? argv[1][0] -
  '0' :
> > 0 );
> >     else if( argc == 2 )
> >         capture = cvCaptureFromAVI( argv[1] );
> >
> >     if( !capture )
> >     {
> >         fprintf(stderr,"Could not initialize capturing...\n");
> >         return -1;
> >     }
> >
> >     cvNamedWindow( "R", CV_WINDOW_AUTOSIZE );
> >     cvSetCaptureProperty(capture,CV_CAP_PROP_FRAME_WIDTH , 640);
> >     cvSetCaptureProperty(capture,CV_CAP_PROP_FRAME_HEIGHT , 480);
> >
> >
> >
> >     for(;;)
> >     {
> >         IplImage* frame = 0;
> >         int i;
> >
> >         frame = cvQueryFrame( capture );
> >         if( !frame )
> >             break;
> >
> >             for( i = 0; i < 3; i++ )
> >                 planes[i] = cvCreateImage( cvSize(frame-
>width,frame-
> > >height), 8, 1 );
> >
> >         cvCvtPixToPlane( frame, planes[0], planes[1], planes[2],
0 );
> >
> > 		    for( i = 0; i < 3; i++ )
> > 		       planes[i]->origin = frame->origin;
> >
> >         cvShowImage("R", planes[0]);
> >
> >         if( cvWaitKey(10) >= 0 )
> >             break;
> >     }
> >
> >     cvReleaseCapture( &capture );
> >     cvDestroyWindow("Video-Window");
> >
> >     return 0;
> > }

I just installed the opencv-0.9.6 release on a Redhat 9 system in my
home directory with

./configure --prefix=$HOME
make
make install

The install seems to go fine, but there are two problems:

1) The test programs cvtest and cxcoretest done't appear in ~/bin

2) When I compile the samples in ~/share/opencv/samples/c with
-L$HOME/lib and set the environment variable LD_LIBRARY_PATH=$HOME/lib
and then try to run them, I get a segfault.  The gdb stack output
looks like this

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 1085764128 (LWP 27633)]
0x405b0f73 in gdk_parse_args () from /usr/lib/libgdk-x11-2.0.so.0
(gdb) bt
#0  0x405b0f73 in gdk_parse_args () from /usr/lib/libgdk-x11-2.0.so.0
#1  0x4041c81f in gtk_parse_args () from /usr/lib/libgtk-x11-2.0.so.0
#2  0x4041cd26 in gtk_init_check () from /usr/lib/libgtk-x11-2.0.so.0
#3  0x4041cd76 in gtk_init () from /usr/lib/libgtk-x11-2.0.so.0
#4  0x401dcb54 in cvNamedWindow (name=0x40355ea4 "l¢", flags=0)
     at window_lnx.cpp:137
#5  0x42015704 in __libc_start_main () from /lib/tls/libc.so.6

Other information:
~$ gcc -v
Reading specs from /usr/lib/gcc-lib/i386-redhat-linux/3.2.2/specs
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man
--infodir=/usr/share/info --enable-shared --enable-threads=posix
--disable-checking --with-system-zlib --enable-__cxa_atexit
--host=i386-redhat-linux
Thread model: posix
gcc version 3.2.2 20030222 (Red Hat Linux 3.2.2-5)

~$ uname -a
Linux cramer-lnx 2.4.20-31.9smp #1 SMP Tue Apr 13 17:40:10 EDT 2004
i686 i686 i386 GNU/Linux


Any ideas?

-Charlie

Hi,
I just saw that there is a new version of OpenCV. The empty test
application that generated the problem works now.

Regards
Andreas Pokorny

On Tue, Aug 31, 2004 at 02:34:47PM +0200, Andreas Pokorny <diemumiee@...>
wrote:
> Hello,
> I ran into problems using OpenCV-0.95.
>
> Here is the output of the debugger:
> Program received signal SIGSEGV, Segmentation fault.
> [Switching to Thread 16384 (LWP 11269)]
> 0x0000002a956b37be in global constructors keyed to icvTextFacesFn0 () at
cvouttext.cpp:655
> 655 cvouttext.cpp: No such file or directory. in cvouttext.cpp
> (gdb) bt
> #0  0x0000002a956b37be in global constructors keyed to icvTextFacesFn0 () at
cvouttext.cpp:655
> #1  0x0000002a957b4c06 in __do_global_ctors_aux () from
/usr/lib/libopencv-0.9.so.5
> #2  0x0000002a9568d573 in _init () from /usr/lib/libopencv-0.9.so.5
> #3  0x0000002a95e8c4b0 in ?? ()
> #4  0x0000002a9556039d in call_init () from /lib64/ld-linux-x86-64.so.2
> #5  0x0000002a9556046b in _dl_init_internal () from
/lib64/ld-linux-x86-64.so.2
> #6  0x0000002a95556ac2 in _dl_start_user () from /lib64/ld-linux-x86-64.so.2
> #7  0x0000000000000001 in ?? ()
> #8  0x0000007fbffff34b in ?? ()
> <-snip->
> up to 500 lines appear here
> <-snap->
> #496 0x7572742f6e657474 in ?? ()
> #497 0x6372756f732f6b6e in ?? ()
> #498 0x616d2f6372732f65 in ?? ()
> #499 0x00747365745f6e69 in ?? ()
> #500 0x0000000000000000 in ?? ()
> Cannot access memory at address 0x7fc0000000
>
> As you can see this seems to happen within the init code of the library.
> The application i play with at the moment is just an empty "int main".
>
> This problem occurs on a linux x86_64 system with
> gcc-version:
> gcc (GCC) 3.4.1 20040803 (Gentoo Linux 3.4.1-r2, ssp-3.4-2, pie-8.7.6.5)
> binutils-2.15.90.0.1.1
> glibc-2.3.4.20040808
>
> Both the application and the library has been built with this setup.
> I am using the output of opencv-config --libs which is '-lopencv -lcvaux
-lhighgui'
> as linker flags.
>
> ldd ./main_test says:
> libopencv-0.9.so.5 => /usr/lib/libopencv-0.9.so.5 (0x0000002a95658000)
> libcvaux-0.9.so.5 => /usr/lib/libcvaux-0.9.so.5 (0x0000002a958e7000)
> libhighgui-0.9.so.5 => /usr/lib/libhighgui-0.9.so.5 (0x0000002a95a7a000)
> libstdc++.so.6 => /usr/lib/gcc-lib/x86_64-pc-linux-gnu/3.4.1/libstdc++.so.6
(0x0000002a95b98000)
> libm.so.6 => /lib/libm.so.6 (0x0000002a95d8a000)
> libgcc_s.so.1 => /usr/lib/gcc-lib/x86_64-pc-linux-gnu/3.4.1/libgcc_s.so.1
(0x0000002a95e8d000)
> libc.so.6 => /lib/libc.so.6 (0x0000002a95f98000)
> libdl.so.2 => /lib/libdl.so.2 (0x0000002a961a3000)
> libpthread.so.0 => /lib/libpthread.so.0 (0x0000002a962a6000)
> libXt.so.6 => /usr/X11R6/lib/libXt.so.6 (0x0000002a9642c000)
> libXm.so.1 => /usr/X11R6/lib/lesstif/libXm.so.1 (0x0000002a9658f000)
> libMrm.so.1 => /usr/X11R6/lib/lesstif/libMrm.so.1 (0x0000002a967db000)
> libUil.so.1 => /usr/X11R6/lib/lesstif/libUil.so.1 (0x0000002a968f2000)
> libpng.so.3 => /usr/lib/libpng.so.3 (0x0000002a96a12000)
> libjpeg.so.62 => /usr/lib/libjpeg.so.62 (0x0000002a96b48000)
> libz.so.1 => /lib/libz.so.1 (0x0000002a96c69000)
> libtiff.so.3 => /usr/lib/libtiff.so.3 (0x0000002a96d7c000)
> libavcodec.so => /usr/lib/libavcodec.so (0x0000002a96ecf000)
> /lib64/ld-linux-x86-64.so.2 => /lib64/ld-linux-x86-64.so.2
(0x0000002a95556000)
> libX11.so.6 => /usr/X11R6/lib/libX11.so.6 (0x0000002a9719f000)
> libSM.so.6 => /usr/X11R6/lib/libSM.so.6 (0x0000002a9737e000)
> libICE.so.6 => /usr/X11R6/lib/libICE.so.6 (0x0000002a97489000)
> libfreetype.so.6 => /usr/lib/libfreetype.so.6 (0x0000002a975a4000)
> libXft.so.2 => /usr/X11R6/lib/libXft.so.2 (0x0000002a9772d000)
> libXrender.so.1 => /usr/X11R6/lib/libXrender.so.1 (0x0000002a97842000)
> libmp3lame.so.0 => /usr/lib/libmp3lame.so.0 (0x0000002a9794b000)
> libvorbis.so.0 => /usr/lib/libvorbis.so.0 (0x0000002a97ae3000)
> libvorbisenc.so.2 => /usr/lib/libvorbisenc.so.2 (0x0000002a97c0f000)
> libXext.so.6 => /usr/X11R6/lib/libXext.so.6 (0x0000002a97ec3000)
> libfontconfig.so.1 => /usr/lib/libfontconfig.so.1 (0x0000002a97fd4000)
> libexpat.so.0 => /usr/lib/libexpat.so.0 (0x0000002a98108000)
> libogg.so.0 => /usr/lib/libogg.so.0 (0x0000002a9822a000)
>
> Any ideas?
>
> Regards
> Andreas Pokorny

Hi Violaine,

Maybe my wording on the blog was a little bit misleading:
"The reason I picked points on these three images is because these
are the images that DEFINES the world coordinate system", I meant
those three IMAGES instead of POINTS :) The extrinsics canNOT be
define by only 3 points in space, it should be defined by a set of
points generally more than 7, in my case 35 points/corners.

So, I only REALIZED the world origin is defined at the upper-left
corner AFTER I tested numerous times by giving THREE 2D-points from
each image (three cameras) then get 3D world coords for that ONE
point in space. The axis are also my "guess" according to the output
3D position of object movement in the live scene since I know how I
actually moved in the space (given that my guess about world origin
is correct).

i know it's very hard to understand everything in OpenCV routines,
you have to even read some utility routines like matrix
multiplication in order to understand how they actually get the
extrinsics, and what the numbers actually mean...

I'm not sure if you are doing 3D tracking stuff, if you are, you
should read the source code besides their documentation(too simple).

Good luck!
mary

--- In OpenCV@yahoogroups.com, "vioty22" <vioty22@y...> wrote:
> Hi Mary,
>
> thank you very much.
> I looked your project but I don't understand why you use lower
left
> and right corners to define world coordinate system. If the upper
> left corner of the chessboard is the translation of world origin,
why
> don't you use upper right corner (axis x) and lower left corner
(axis
> y). Can you explain me how you calculate the world coordinate
system
> with these 3 points.
>
> Bye,
> Violaine
>
>
> --- In OpenCV@yahoogroups.com, "Mary" <molobu2001@y...> wrote:
> > Hi Violanine,
> >
> > I think the vector represents the translation of world origin
> > (0,0,0) which is at the upper-left most inner cornner of your
chess
> > board.
> >
> > I had tested my calibration results by giving the pixel image
> coords
> > of the upper-left corner to my tracking system, instead of
(0,0,0),
> > it returns me something like (18.5, 18.5, 19.0) unit in "cm". I
> > think it's a torlerable result.
> >
> > you can check out more details on my project blog if you are
> > interested:
> > http://www.engineering.ucsb.edu/~mmllee/research/prgs0906.html
> >
> > good luck!
> > mary
> >
> > --- In OpenCV@yahoogroups.com, "vioty22" <vioty22@y...> wrote:
> > > Hello,
> > >
> > > I just want to know if someone use this function
> > > (cvFindExtrinsicCameraParams) and if he can explain me what
> > represent
> > > exactly the matrix and vector returned. I think that the
matrix
> is
> > > the matrix of rotation which gives orientation of chessboard
in
> > > camera locate. For this vector, I think it's the position of a
> > > chessboard inner corner but which one ?
> > >
> > > Thanks a lot in advance.
> > > Bye
> > >
> > > Violaine

Hi Everybody
I've been subscribed to this list for a while now and I find there's
quite a bit of spam coming from here. Can anything be done?
(Of course I have a working spam-filter so it's not that big a problem,
but anyway...)

Søren

It seems that CvBox2D::angle stores the degree value of the angle
instead
of Radians, but in the document
it says that it is a Radian value. I guess that's why the function
cvBoxPoints() returns wrong value since it
takes sin() and cos() directly on the degree value of the angle.

Best,

Gang Hua

Glad it's not just me, I made the same discovery this
morning.

-ejay

...

Bad:
for (;;) {
  ...Capture Image...
  gray=cvCreateImage(image->width, image->height...);
  gray=cvCvtColor(.image, gray, CV_BGR2GRAY);
  .Do something..
}

Good:
...Capture Image...
  gray=cvCreateImage(image->width, image->height...)

for (;;) {
...Capture Image...
  gray=cvCvtColor(.image, gray, CV_BGR2GRAY);
  .Do something..
}




> -----Original Message-----
> From: tgeorgio [mailto:tristen_georgiou@...]
> Sent: Tuesday, August 31, 2004 2:35 PM
> To: OpenCV@yahoogroups.com
> Subject: [OpenCV] Re: Memory Issue
>
> I found the problem...I assumed that when loading/updating
an image
> using the IplImage structure, that it would overwrite the
existing
> imagedata, but what I think it does is just move the
imagedata
> pointer to a free spot in memory (w/o clearing the
existing memory).
> Since I re-use a few of the IplImage types, I found that I
would have
> to make a call to cvReleaseData before loading/updating
the image.
> This fixed my problem immediately.  I am not sure if this
is the
> correct way to do it (should I maybe call cvReleaseImage
instead to
> clear the header too?), but it has put a stop to my major
memory leak.
>
>
> --- In OpenCV@yahoogroups.com, "tnyx1" <xut@k...> wrote:
> > Do you use CvMemStorage?  If yes, you should call
cvClearMemStorage
> > after each processing cycle.
> >
> >
> > --- In OpenCV@yahoogroups.com, "tgeorgio"
<tristen_georgiou@h...>
> > wrote:
> > > I have noticed that when my application is running, it
starts
> eating
> > > memory like crazy, eventually using up all free memory
as well as
> > swap
> > > space, and bringing my computer to a standstill.  I
was pretty
> > careful
> > > about releasing memory I was no longer using, and when
I shut my
> > > program down, all of this memory is returned to the OS
(I do a
> > cleanup
> > > upon shutdown).  I am mostly just using the IplImage
structure;
> do I
> > > need to release and then reallocate this structure
every time
> > before I
> > > load in another image (by cvCloneImage, from camera or
from
> file)?
> > > Should I run my clean up function periodically?
> > >
> > > I probably have just forgotten to release some memory
somewhere
> but
> > > any suggestions on how to make my program more memory
efficient
> > would
> > > be appreciated.  Thanks,
> > >
> > > Tristen
>
>
>
> ------------------------ Yahoo! Groups Sponsor
> --------------------~-->
> Make a clean sweep of pop-up ads. Yahoo! Companion
Toolbar.
> Now with Pop-Up Blocker. Get it for free!
> http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/W4wwlB/TM
>
------------------------------------------------------------
--
> ------~->
>
> Change settings: http://www.yahoogroups.com/mygroups,
select
>    Get Emails (get all posts)
>    Daily Digest (one summary email per day)
>    Read on the web (read posts on the web only)Or
Unsubscribe
> by mailing OpenCV-unsubscribe@yahoogroups.com
>
> Yahoo! Groups Links
>
>
>
>

I too have the same problem..

mine is at stage 10..  but have had it happen on stage 18 too so i'm
not sure whats going on

--- In OpenCV@yahoogroups.com, Zhanfeng Yue <yzf1973@y...> wrote:
> I have the similar problem but it is at stage 7.
> Anyone can help?
> --- subsins <subsins@y...> wrote:
>
> > hye all
> > i think beta 4 got some problems with in cause i get
> > memory could not
> > be written at stage 0 can anyone point to this
> > problem what is the
> > problem with haartraining.
> > thanks in advance,
> > khuram.
> >
> >
> >
> > ------------------------ Yahoo! Groups Sponsor
> > --------------------~-->
> > Make a clean sweep of pop-up ads. Yahoo! Companion
> > Toolbar.
> > Now with Pop-Up Blocker. Get it for free!
> >
> http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/W4wwlB/TM
> >
> --------------------------------------------------------------------~->
> >
> >
> > Change settings:
> > http://www.yahoogroups.com/mygroups, select
> >    Get Emails (get all posts)
> >    Daily Digest (one summary email per day)
> >    Read on the web (read posts on the web only)Or
> > Unsubscribe by mailing
> > OpenCV-unsubscribe@yahoogroups.com
> >
> > Yahoo! Groups Links
> >
> >
> >     OpenCV-unsubscribe@yahoogroups.com
> >
> >
> >
> >
>
>
>
>
> __________________________________
> Do you Yahoo!?
> Y! Messenger - Communicate in real time. Download now.
> http://messenger.yahoo.com

Hi all,

I am getting the following error:

OpenCV Error: Bad/unsupported parameter of type CvSize
         in function [cvsmooth.cpp:2984]:cvSmooth():
OpenCV function failed
OpenCV: terminating the application

Here is the code that is causing the problem:

IplImage* FindGaussianDifference(IplImage* img)
{
   IplImage* vertGaussian;
   IplImage* horzGaussian;
   IplImage* result;
   CvSize imgSize;

   // result size is the same as loaded image size
   imgSize.width = img->width;
   imgSize.height = img->height;

   vertGaussian = cvCreateImage( imgSize, img->depth, img->nChannels );
   horzGaussian = cvCreateImage( imgSize, img->depth, img->nChannels );
   result = cvCreateImage( imgSize, img->depth, img->nChannels );

   // Perform the gaussian smoothing in the horizontal and vertical
directions
   cvSmooth( img, vertGaussian, CV_GAUSSIAN, 3, 0 );
   cvSmooth( img, horzGaussian, CV_GAUSSIAN, 0, 3 );

   // find the difference of the gaussian transforms
   cvSub( vertGaussian, horzGaussian, result, 0);

   return result;
}

Can anyone tell me what I am doing wrong?

Thanks so much for any assitance you can provide,
James Fitzsimons

--
It's 5.50 a.m.... Do you know where your stack pointer is ?

Hi everyone!

Currently, I am using HighGUI to capture avi video file, it works
very well in Visual C++ 6.0, but when I tried to incorporate it
into Visual C++ .Net framework, the highgui.h produces many compiling
error. Does anybody know what the problem is? How to overcome it?

I'd really appreciate for your helping~~~

Best,
Ting

hi
i want to use openCV in VC++.net or Embeded VC++.


functions of OpenCV could be used in Embeded VC++,or VC++.net which
runs in wince?

Hey..

does anyone know what

"Clusters are too small. Clustering aborted."  means?

I'm getting this message while training a classifier.

Any ideas?

Cheers!
  - ROb

Not sure.. i've notice tons of spam too..

especially from some sender 'vancouver Webpages Web gateway'


--- In OpenCV@yahoogroups.com, Soren Hauberg <soren@h...> wrote:
> Hi Everybody
> I've been subscribed to this list for a while now and I find there's
> quite a bit of spam coming from here. Can anything be done?
> (Of course I have a working spam-filter so it's not that big a problem,
> but anyway...)
>
> Søren

Bad => Good
for (;;) {
 ...Capture Image...
 gray=cvCreateImage(image->width, image->height...);

 //gray=cvCvtColor(.image, gray, CV_BGR2GRAY);  // cvCvtColor doesn't return anything

cvCvtColor(image, gray, CV_BGR2GRAY);

 .Do something..

  cvReleaseImage( &gray );
}

Good => Better
...Capture Image...
 gray=cvCreateImage(image->width, image->height...)

for (;;) {
...Capture Image...

cvZero(gray);

// gray=cvCvtColor(.image, gray, CV_BGR2GRAY); // cvCvtColor doesnt return anything

cvCvtColor(image, gray, CV_BGR2GRAY);

 .Do something..
}
 

I hope this solves your problem. Revert for any further queries;
Saurabh
 

Ejay Hire wrote:

Glad it's not just me, I made the same discovery this
morning.

-ejay

...

Bad:
for (;;) {
 ...Capture Image...
 gray=cvCreateImage(image->width, image->height...);
 gray=cvCvtColor(.image, gray, CV_BGR2GRAY);
 .Do something..
}

Good:
...Capture Image...
 gray=cvCreateImage(image->width, image->height...)

for (;;) {
...Capture Image...
 gray=cvCvtColor(.image, gray, CV_BGR2GRAY);
 .Do something..
}

> -----Original Message-----
> From: tgeorgio [mailto:tristen_georgiou@...]
> Sent: Tuesday, August 31, 2004 2:35 PM
> To: OpenCV@yahoogroups.com
> Subject: [OpenCV] Re: Memory Issue
>
> I found the problem...I assumed that when loading/updating
an image
> using the IplImage structure, that it would overwrite the
existing
> imagedata, but what I think it does is just move the
imagedata
> pointer to a free spot in memory (w/o clearing the
existing memory).
> Since I re-use a few of the IplImage types, I found that I
would have
> to make a call to cvReleaseData before loading/updating
the image.
> This fixed my problem immediately.  I am not sure if this
is the
> correct way to do it (should I maybe call cvReleaseImage
instead to
> clear the header too?), but it has put a stop to my major
memory leak.
>
>
> --- In OpenCV@yahoogroups.com, "tnyx1" <xut@k...> wrote:
> > Do you use CvMemStorage?  If yes, you should call
cvClearMemStorage
> > after each processing cycle.
> >
> >
> > --- In OpenCV@yahoogroups.com, "tgeorgio"
<tristen_georgiou@h...>
> > wrote:
> > > I have noticed that when my application is running, it
starts
> eating
> > > memory like crazy, eventually using up all free memory
as well as
> > swap
> > > space, and bringing my computer to a standstill.  I
was pretty
> > careful
> > > about releasing memory I was no longer using, and when
I shut my
> > > program down, all of this memory is returned to the OS
(I do a
> > cleanup
> > > upon shutdown).  I am mostly just using the IplImage
structure;
> do I
> > > need to release and then reallocate this structure
every time
> > before I
> > > load in another image (by cvCloneImage, from camera or
from
> file)?
> > > Should I run my clean up function periodically?
> > >
> > > I probably have just forgotten to release some memory
somewhere
> but
> > > any suggestions on how to make my program more memory
efficient
> > would
> > > be appreciated.  Thanks,
> > >
> > > Tristen
>
>
>
> ------------------------ Yahoo! Groups Sponsor
> --------------------~-->
> Make a clean sweep of pop-up ads. Yahoo! Companion
Toolbar.
> Now with Pop-Up Blocker. Get it for free!
> http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/W4wwlB/TM
>
------------------------------------------------------------
--
> ------~->
>
> Change settings: http://www.yahoogroups.com/mygroups,
select
>    Get Emails (get all posts)
>    Daily Digest (one summary email per day)
>    Read on the web (read posts on the web only)Or
Unsubscribe
> by mailing OpenCV-unsubscribe@yahoogroups.com
>
> Yahoo! Groups Links
>
>
>
>

------------------------ Yahoo! Groups Sponsor --------------------~-->
$9.95 domain names from Yahoo!. Register anything.
http://us.click.yahoo.com/J8kdrA/y20IAA/yQLSAA/W4wwlB/TM
--------------------------------------------------------------------~->

Change settings: http://www.yahoogroups.com/mygroups, select
   Get Emails (get all posts)
   Daily Digest (one summary email per day)
   Read on the web (read posts on the web only)Or Unsubscribe by mailing OpenCV-unsubscribe@yahoogroups.com

Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/OpenCV/

<*> To unsubscribe from this group, send an email to:
    OpenCV-unsubscribe@yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/
 

I've just banned it.

--- In OpenCV@yahoogroups.com, "crypticvoid" <yahoo@t...> wrote:
> Not sure.. i've notice tons of spam too..
>
> especially from some sender 'vancouver Webpages Web gateway'
>
>
> --- In OpenCV@yahoogroups.com, Soren Hauberg <soren@h...> wrote:
> > Hi Everybody
> > I've been subscribed to this list for a while now and I find there's
> > quite a bit of spam coming from here. Can anything be done?
> > (Of course I have a working spam-filter so it's not that big a
problem,
> > but anyway...)
> >
> > Søren

Max,

In OpenCV workspace there is a project "cvsample".
Currently it contains morphology.c. You can exclude morphology.c
and add facedetect.c (or any other sample).
Note, that you have to set proper working directory and
add --cascade="<path to cascade>" to the command-line options.

On the profile face detection.
As I understood, training the cascade on both left and right profiles
makes the recognition performance (quality) much worse.

As a workaround, you may run cvHaarDetectObjects twice - on original
and horizontally (cvFlip(img,img,1)) flipped image.
Of course, in the latter case you'd have to flip the output rectangles.

Vadim

--- In OpenCV@yahoogroups.com, "Massimo De Vivo" <m.devivo@t...> wrote:
> Hi all,
> in the beta4 release, haarfacedetect sample doesn't exist anymore and
> the only sample for haar face detection is facedetect. I've found the .c
> file, but I cannot find any workspace or project, which includes it.
> Does it exist or do I have to create it by myself?
> Just another question: I've tried the profile classifier and it seems to
> work fine, but it's only for left profile, so for the right profile I
> have to flip the image and then apply the classifier. But I can guess
> and I know that is possible to vary the classifier and not the image in
> order to detect right profile. I mean "to flip the classifier", varying
> the parameters and not the image, that is faster. Has anyone idea of how
> to do this? How do I have to vary haar parameters in such a way to
> detect flipped objects?
>
> Thanks and regards,
> Max

Hi,
I've tested cvcap_v4l.cpp and it has solved all the problems I've
been relating previously.
My Creative Labs Webcam III works fine now.

Congratulations for the professional work of the group.

Thankx
Carlos

--- In OpenCV@yahoogroups.com, "Pisarevsky, Vadim"
<Vadim.Pisarevsky@i...> wrote:
> Hello all OpenCV on Linux users,
>
>
>
> Improved (or completely rewritten) versions of
>
> cvcap_v4l.cpp (by Sfuncia Fabio)
>
> and cvcap_dc1394.cpp (by Frederic Devernay)
>
> that add support for more cameras are available on CVS.
>
> With updated configure.in different versions of libdc1394 are now
> supported
>
> (another patch from Frederic).
>
>
>
> Vadim

I want to use CONDENSATION algorithm in OpenCV to track object. But I
don't know how to do it. I didn't get understand for the sample code
of CONDENSATION algorithm in OpenCV.

Who can help me to make clear how to use functions of CONDENSATION in
OpenCV? With the best, could you supply some simple and pellucid
sample code for me.

Thank you very much!

Expect your answer eagerly!

Hi Florian,

You may use cvLoadHaarClassifierCascade() to load cascade from
a folder.
The loaded cascades may be saved to XML using cvSave() function.

I put samples/c/convert_cascade.c to OpenCV CVS at SourceForge.
It can do the job.

Vadim

--- In OpenCV@yahoogroups.com, "Florian" <robotik@i...> wrote:
> Hi,
> I wonder how I can convert old cascades (textfiles in directories) into
> the format introduced in beta 4 (xml file)?
> Thx for any pointers
> --Florian

Hi,
  I also put a small visual studio project called "haarconv" under "Files
> Examples"
(http://groups.yahoo.com/group/OpenCV/files/Examples/haarconv.zip)
Thx
--Florian

-----Original Message-----
From: vp153 [mailto:Vadim.Pisarevsky@...]
Sent: Friday, September 03, 2004 10:05 AM
To: OpenCV@yahoogroups.com
Subject: [OpenCV] Re: OR classifier conversion


Hi Florian,

You may use cvLoadHaarClassifierCascade() to load cascade from
a folder.
The loaded cascades may be saved to XML using cvSave() function.

I put samples/c/convert_cascade.c to OpenCV CVS at SourceForge.
It can do the job.

Vadim

--- In OpenCV@yahoogroups.com, "Florian" <robotik@i...> wrote:
> Hi,
> I wonder how I can convert old cascades (textfiles in directories)
into
> the format introduced in beta 4 (xml file)?
> Thx for any pointers
> --Florian






Change settings: http://www.yahoogroups.com/mygroups, select
    Get Emails (get all posts)
    Daily Digest (one summary email per day)
    Read on the web (read posts on the web only)Or Unsubscribe by mailing
OpenCV-unsubscribe@yahoogroups.com

Yahoo! Groups Links

Thanks

vp153 wrote:
> I've just banned it.
>
> --- In OpenCV@yahoogroups.com, "crypticvoid" <yahoo@t...> wrote:
>
>>Not sure.. i've notice tons of spam too..
>>
>>especially from some sender 'vancouver Webpages Web gateway'
>>
>>
>>--- In OpenCV@yahoogroups.com, Soren Hauberg <soren@h...> wrote:
>>
>>>Hi Everybody
>>>I've been subscribed to this list for a while now and I find there's
>>>quite a bit of spam coming from here. Can anything be done?
>>>(Of course I have a working spam-filter so it's not that big a
>
> problem,
>
>>>but anyway...)
>>>
>>>Søren
>
>
>
>
>
> Change settings: http://www.yahoogroups.com/mygroups, select
>    Get Emails (get all posts)
>    Daily Digest (one summary email per day)
>    Read on the web (read posts on the web only)Or Unsubscribe by mailing
OpenCV-unsubscribe@yahoogroups.com
>
> Yahoo! Groups Links
>
>
>
>
>
>

Hello.
I also worry about the same thing(get 640x480 web cam window).

The development environment is as follows.
OS:Windows2000
OpenCV:opencv-win(beta4)
Softwere:Visual C++ .NET 2003
Camera:Logitech QuickCam Pro 4000

Which file is rewritten how? Then, what thing should be done?
(build and dll copy to SYSTEM32?)

If it avoids, please give me advice.

dandelion


--- In OpenCV@yahoogroups.com, "juanwachs" <juan@b...> wrote:
> Well! It worked also for the windows xp.
> I follow your instructions and it works!
> Thanks!
> Juan
>
>
> -- In OpenCV@yahoogroups.com, "tkwoodfamily" <yahoo@k...> wrote:
> > Linux or Windows?
> >
> > Linux Beta4? Go to file:
> > /opencv/otherlibs/highgui/cvcap_v4l.cpp
> > take out the double slash (//) that will un comment the line:
> > #define V4L_SCALE_OPTION 1
> > also change the DEFAULT_WIDTH and DEFAULT_HEIGHT to your prefered
> size.
> > then -
> >
> > make
> > make install
> >
> > Re run and enjoy
> >
> >
> > Windows Beta4?  Dont know.  Someone else will have to help you.
> >
> > Travis Wood
> >
> >
> >
> >
> > --- In OpenCV@yahoogroups.com, "juanwachs" <juan@b...> wrote:
> > > Dear OpenCV users:
> > > I'm trying to modify the following code to increase the window
of
> the
> > > web-cam to 640x480. I know that there is an application
> > > called 'cvcamdemo' inside the cvcam folder of the OpenCV, which
> > > change the window size through a program, but I have problems
> adding
> > > new windows there.
> > > So I prefer to change the following code. Why the commands
> > >    cvSetCaptureProperty(capture,CV_CAP_PROP_FRAME_WIDTH ,
640)
> and
> > >     cvSetCaptureProperty(capture,CV_CAP_PROP_FRAME_HEIGHT ,
480);
> > > are not working???
> > > Need help!
> > >
> > > int main(int argc, char* argv[])
> > > {
> > >  int out=0;
> > >  IplImage* laplace = 0;
> > >     IplImage* colorlaplace = 0;
> > >     IplImage* planes[3] = { 0, 0, 0 };
> > >     CvCapture* capture = 0;
> > >
> > >
> > >     if( argc == 1 || (argc == 2 && strlen(argv[1]) == 1 &&
isdigit
> > > (argv[1][0])))
> > >         capture = cvCaptureFromCAM( argc == 2 ? argv[1][0] -
>  '0' :
> > > 0 );
> > >     else if( argc == 2 )
> > >         capture = cvCaptureFromAVI( argv[1] );
> > >
> > >     if( !capture )
> > >     {
> > >         fprintf(stderr,"Could not initialize capturing...\n");
> > >         return -1;
> > >     }
> > >
> > >     cvNamedWindow( "R", CV_WINDOW_AUTOSIZE );
> > >     cvSetCaptureProperty(capture,CV_CAP_PROP_FRAME_WIDTH , 640);
> > >     cvSetCaptureProperty(capture,CV_CAP_PROP_FRAME_HEIGHT ,
480);
> > >
> > >
> > >
> > >     for(;;)
> > >     {
> > >         IplImage* frame = 0;
> > >         int i;
> > >
> > >         frame = cvQueryFrame( capture );
> > >         if( !frame )
> > >             break;
> > >
> > >             for( i = 0; i < 3; i++ )
> > >                 planes[i] = cvCreateImage( cvSize(frame-
> >width,frame-
> > > >height), 8, 1 );
> > >
> > >         cvCvtPixToPlane( frame, planes[0], planes[1], planes
[2],
> 0 );
> > >
> > > 		    for( i = 0; i < 3; i++ )
> > > 		       planes[i]->origin = frame->origin;
> > >
> > >         cvShowImage("R", planes[0]);
> > >
> > >         if( cvWaitKey(10) >= 0 )
> > >             break;
> > >     }
> > >
> > >     cvReleaseCapture( &capture );
> > >     cvDestroyWindow("Video-Window");
> > >
> > >     return 0;
> > > }

--- In OpenCV@yahoogroups.com, "sbrubaker6" <scb@s...> wrote:
> I just installed the opencv-0.9.6 release on a Redhat 9 system in my
> home directory with
>
> ./configure --prefix=$HOME
> make
> make install
>
> The install seems to go fine, but there are two problems:
>
> 1) The test programs cvtest and cxcoretest done't appear in ~/bin
>
> 2) When I compile the samples in ~/share/opencv/samples/c with
> -L$HOME/lib and set the environment variable LD_LIBRARY_PATH=$HOME/lib
> and then try to run them, I get a segfault.  The gdb stack output
> looks like this
>
> Program received signal SIGSEGV, Segmentation fault.
> [Switching to Thread 1085764128 (LWP 27633)]
> 0x405b0f73 in gdk_parse_args () from /usr/lib/libgdk-x11-2.0.so.0
> (gdb) bt
> #0  0x405b0f73 in gdk_parse_args () from /usr/lib/libgdk-x11-2.0.so.0
> #1  0x4041c81f in gtk_parse_args () from /usr/lib/libgtk-x11-2.0.so.0
> #2  0x4041cd26 in gtk_init_check () from /usr/lib/libgtk-x11-2.0.so.0
> #3  0x4041cd76 in gtk_init () from /usr/lib/libgtk-x11-2.0.so.0
> #4  0x401dcb54 in cvNamedWindow (name=0x40355ea4 "l¢", flags=0)
>     at window_lnx.cpp:137
> #5  0x42015704 in __libc_start_main () from /lib/tls/libc.so.6
>
> ...
> -Charlie

Charlie,

try to replace line 204 in otherlibs/highgui/window_lnx.cpp
cvInitSystem(0,0);
with
cvInitSystem(1,(char**)name);

and rebuild highgui.
Probably, it will resolve the problem.

cxcoretest and cvtest are not installed intentionally
(so, INSTALL document is incorrect). You may run the tests
from their build directories.

Vadim

Oops, the line should look:
cvInitSystem(1,(char**)&name); // note &

Vadim

--- In OpenCV@yahoogroups.com, "vp153" <Vadim.Pisarevsky@i...> wrote:
> --- In OpenCV@yahoogroups.com, "sbrubaker6" <scb@s...> wrote:
> > I just installed the opencv-0.9.6 release on a Redhat 9 system in my
> > home directory with
> >
> > ./configure --prefix=$HOME
> > make
> > make install
> >
> > The install seems to go fine, but there are two problems:
> >
> > 1) The test programs cvtest and cxcoretest done't appear in ~/bin
> >
> > 2) When I compile the samples in ~/share/opencv/samples/c with
> > -L$HOME/lib and set the environment variable LD_LIBRARY_PATH=$HOME/lib
> > and then try to run them, I get a segfault.  The gdb stack output
> > looks like this
> >
> > Program received signal SIGSEGV, Segmentation fault.
> > [Switching to Thread 1085764128 (LWP 27633)]
> > 0x405b0f73 in gdk_parse_args () from /usr/lib/libgdk-x11-2.0.so.0
> > (gdb) bt
> > #0  0x405b0f73 in gdk_parse_args () from /usr/lib/libgdk-x11-2.0.so.0
> > #1  0x4041c81f in gtk_parse_args () from /usr/lib/libgtk-x11-2.0.so.0
> > #2  0x4041cd26 in gtk_init_check () from /usr/lib/libgtk-x11-2.0.so.0
> > #3  0x4041cd76 in gtk_init () from /usr/lib/libgtk-x11-2.0.so.0
> > #4  0x401dcb54 in cvNamedWindow (name=0x40355ea4 "l¢", flags=0)
> >     at window_lnx.cpp:137
> > #5  0x42015704 in __libc_start_main () from /lib/tls/libc.so.6
> >
> > ...
> > -Charlie
>
> Charlie,
>
> try to replace line 204 in otherlibs/highgui/window_lnx.cpp
> cvInitSystem(0,0);
> with
> cvInitSystem(1,(char**)name);
>
> and rebuild highgui.
> Probably, it will resolve the problem.
>
> cxcoretest and cvtest are not installed intentionally
> (so, INSTALL document is incorrect). You may run the tests
> from their build directories.
>
> Vadim

Hi,

I am trying to get a PAL-camera to work with OpenCv via an Canopus
ADVC-55 converter. This connects to ieee1394. By using the
dv1394-driver and dvgrab tool Iwas able to grab an avi from the cam.
Due to the necessity of image processing I want to use this cam with
OpenCv. But I was not able to initialize a CvCapture wether from a
grabbed avi-file nor from the camera directly.

I also tried the the libcam1394 library, but got always a "no cam
found" error.

Does anyone have a sample application for OpenCv with ieee1394-cam?
I'd be glad for any help.

Thanks,

Sami

I'm writing to see if we might be misunderstanding the data which is being
returned from the cvCalibrateCamera method.

We have a wide angle lense with significant distortion in it and we decided
to try to calibrate it using two approaches - the executable Zhang provides
from his website (which includes a skew term in the intrinsic matrix but no
tangential distortion correction term) and the OpenCV cvCalibrateCamera
method (which includes the tangential correction term but no skew term).

When we run the two methods on the same set of data (about 7 images,
markers not in a regular checkerboard pattern and so the detection is done
independently), we get very different answers for the distortion and
intrinsic parameters. For example, Zhang's program gives a horizontal field
of view of 187 pixels, whereas cvCalibrateCamera suggests it's 26.

When I run cvUndistort using the figures from both methods, the results
appear to be very good - very little distortion is evident. However, if I
then try to project a point from a 3D model onto the pixel plane, the
intrinsic parameters from Zhang's program appear to be correct whereas
those from OpenCV appear to be significantly in error.

Therefore, I was wondering if we are we interpreting the output from
cvCalibrateCamera correctly? We assume that the cameraMatrix contains the
intrinsic parameters directly and no scaling is required. Is this
assumption correct? If not, what should we do? (The documentation seems
rather vague.) Is there any way to tell if the calibration iteration failed
to converge? Has anybody else compared cvCalibrateCamera to Zhang's program
and, if so, what results have they encountered?

Many thanks,

Simon Julier

I read haartraining howto but i can't still understand how to pass
negative images to opencv-haartraining.
In my case i have ~/pos/  dir containing file positive.vec created by
opencv-createsamples and positive images, and ~/neg/ dir containin neg
images with file train.txt made as follows:


B1_00001.pgm 1 0 0 19 19
B1_00002.pgm 1 0 0 19 19
B1_00003.pgm 1 0 0 19 19
B1_00004.pgm 1 0 0 19 19
B1_00005.pgm 1 0 0 19 19
B1_00006.pgm 1 0 0 19 19
B1_00007.pgm 1 0 0 19 19
B1_00008.pgm 1 0 0 19 19
B1_00009.pgm 1 0 0 19 19
B1_00010.pgm 1 0 0 19 19
B1_00011.pgm 1 0 0 19 19
.....


Now i launched in ~/pos/ dir

opencv-haartraining -data
~/WORKING/Intel-AVSR/opencv-0.9.6/data/haarcascades/ -vec
positives.vec -bg ~/neg/train.txt


But i  get: "Invalid background description file."

Could anyone explain me my mistake.
Thank you very much, sorry for the stupid question.

Alberto

--- In OpenCV@yahoogroups.com, "tkwoodfamily" <yahoo@k...> wrote:
> Check Out:
> http://inflomatik.com/download/OpenCV_ObjectDetection_HowTo.pdf
> and
> http://www.itu.dk/stud/ projects_f2004/haartracking/OpenCV/apps/
> HaarTracking/doc/haartraining.doc
> For some good examples
>
> Travis Wood
>
> --- In OpenCV@yahoogroups.com, "yuting_nwu" <tingyu@e...> wrote:
> > Does anybody know how to write the <vec_file_name> and
> > <background_file_name> as in the arguments of HaarTraining
> program?
> > I don't know how to write these DAT files in the correct way.
> >
> > For example, I have a set of positive training images named as
> > "pos_01.jpg, pos_02.jpg ..." in the root of hard disk C. and
> > Each image contains a single human face at the location where
> > the left top corner of the face rect is 10 10, and the size of
> > this rect is 24 24, could I simply write the <vec_file_name> as
> > follows?
> >
> > pos_01.jpg 1 10 10 24 24
> > pos_02.jpg 1 10 10 24 24
> > pos_03.jpg 1 10 10 24 24
> > ......
> >
> > And all of the above information are saved in a file, say we name
> > it as "Positive.txt"???
> >
> > Is this the correct way to write this descripition file?
> > But I did like this, and the HaarTraining program seems not
> > to accept this information format. Did I make any mistakes???
> >
> > Anyone can help? Thank you very much!!!
> >
> > Best,
> > Ting