Well First what i can say about the program is this is under GPL
and ...
and those segmentation fault is caused due to unavailability of
(A.txt,B.txt,...Z.txt) files,that it would need to compare it would be
better to get all those files from ( ). This recognition is pretty
cool if you uses same method of writing i mean stroke should be under
law ... for 'B' draw straight line from bottom to top and then from
the top draw 3 like curve to make 'B' press recognise and it's all,
you can easily change this to recognise as your wished stroked
handwriting. Even in some minutes you can port that to any language
lipi, nepali english and so on. I need some tools to check the
performance and compare out it from those general method of
recognition like neural networks. I hope you all guys will co-ordinate
to make this better, might be the best handwriting recognition ever.
what i believe is this can be faster that neural networks technique if
some improvements are given.
Regards
Ujwal
On May 20, 12:03 pm, "nepbabu.cx" <[EMAIL PROTECTED]> wrote:
> [EMAIL PROTECTED] wrote:
> > Ujwal, this one is impressive using GTK library. That's really great
> > one.
>
> Hello dibyendra ju! :P
>
>
>
>
>
> > On May 19, 9:12 am, "Ujwal (RUBBOT)" <[EMAIL PROTECTED]> wrote:
> >> #include <gtk/gtk.h>
> >> #include <stdio.h>
> >> #include <malloc.h>
> >> #include <math.h>
> >> #define part 10
> >> #define chno 26
> >> static int len,count,tm[2][16];
> >> static GdkPixmap *pixmap = NULL;
> >> float statistic_data(char *src, char *des) // Real
> >> Core Code
> >> {
> >> int i=0,j=0,ax,ay,bx,by,sigx=0,sigy=0;
> >> float
> >> xbar,ybar,sigu,sigv=0,sigu2=0,sigv2=0,siguv=0,karl_pearson,karl_pearsony;
> >> float *u,*v,*u2,*v2,*uv;
> >> FILE *X,*Y;
>
> [snip]
>
> Ujwal, I had a quick compile and ran it but I don't know whether it's
> only me or for someone else as well, it's segfaulting on me. I ran a
> backtrace and the culprit seems to be fscanf() in your statistic_data(..).
>
> 20: sigx=sigx+ax; // calculate sum of
> all x
> 21: fscanf(Y,"%d %d ",&bx,&by); // "
>
> == bt below ====
> [Thread debugging using libthread_db enabled]
> [New Thread -1218214224 (LWP 1326)]
>
> 73 is the length
>
> Program received signal SIGSEGV, Segmentation fault.
> [Switching to Thread -1218214224 (LWP 1326)]
> 0xb776557a in _IO_vfscanf () from /lib/tls/i686/cmov/libc.so.6
> (gdb) bt
> #0 0xb776557a in _IO_vfscanf () from /lib/tls/i686/cmov/libc.so.6
> #1 0xb776ace3 in vfscanf () from /lib/tls/i686/cmov/libc.so.6
> #2 0xb776fb72 in fscanf () from /lib/tls/i686/cmov/libc.so.6
> #3 0x0804910f in statistic_data (src=0x804a3a8 "ujwal.txt",
> des=0x804a29d "A.txt") at handwriting.c:21
> #4 0x08049a1a in getxy (area=0x806f328, data=0x808d0f0) at
> handwriting.c:154
> #5 0xb792287b in IA__g_cclosure_marshal_VOID__VOID (closure=0x804c0d8,
> return_value=0x0, n_param_values=1, param_values=0xbfbf913c,
> invocation_hint=0xbfbf904c, marshal_data=0x80498a5) at gmarshal.c:77
> #6 0xb791603b in IA__g_closure_invoke (closure=0x804c0d8,
> return_value=0x0, n_param_values=1, param_values=0xbfbf913c,
> invocation_hint=0xbfbf904c) at gclosure.c:490
> =======
>
> Btw, great work bro! ;)
>
> Cheers
--~--~---------~--~----~------------~-------~--~----~
FOSS Nepal mailing list
[email protected]
http://groups.google.com/group/foss-nepal
Community website: http://www.fossnepal.org/
-~----------~----~----~----~------~----~------~--~---
