Example 21d: FE geometric fitting to laser scanned bone data

This example covers:

• creation of an initial mesh comprised of linear Lagrange elements using nodes obtained using our Faro Arm.
• acquisition of data points on the surface of a cast model of a human long bone (the femur) using the 3Space® FastSCAN™ Hand Held Laser Scanner.
• fitting of the initial mesh to this surface data using cubic Hermite basis functions.

FaroInitialisationFile
serial_port 2
baud_rate 9600
_D1
        

• The best utility for conversion between front and back end node and data formats is cmConvert.Typing cmConvert on the command line will provide help on this utility.
• You will also need to add `versions' to your mesh. You may find the following script useful just to get the overall ipnode format right but will still need to manually edit the file in order to add more than one version.
  

  #!/usr/bin/perl -i.nov
  #Description: Takes an .ipnode file not containing versions and adds a  
  #default one version for each node in nj=1, nj=2, nj=3. Also asks for
  #prompting for different versions.
  #usage: addversions.pl file.ipnode
  
  while (<>) {
   if (/Do you want prompting for different/) {
     s/ N/ Y/;
   }
   if (/The Xj.1./) {
                  s/The Xj.1./The number of versions for nj=1 is [1]: 1\n The Xj(1)/;
          }
   if (/The Xj.2./) {
     s/The Xj.2./The number of versions for nj=2 is [1]: 1\n The Xj(2)/;
   }
   if (/The Xj.3./) {
     s/The Xj.3./The number of versions for nj=3 is [1]: 1\n The Xj(3)/;
   }
          print;
  } 
            
            

• I also use the following for adding derivatives.
  

  #!/usr/bin/perl -i.old
  
  while (<>) {
          if (/derivatives/) {
                  s/: 0/: 3/;
          }
          print;
          if (/Xj/) {
                  print <<EOF;
   The derivative wrt direction 1 is [ 0.00000D+00]: 0.00000D+00
   The derivative wrt direction 2 is [ 0.00000D+00]: 0.00000D+00
   The derivative wrt directions 1 & 2 is [ 0.00000D+00]: 0.00000D+00
  EOF
          }
  } 
            

The comfile run by this example is as follows:

$MULTIPLEFITS=0

{
  eval { require Time::HiRes };
  if( $@ )
    {
      print "warning: $@";
    }
  else
    {
      import Time::HiRes qw(time);
    }
}

sub cpu_time()
  {
    # Add user/system parent/child cpu_times.
    my $result = 0;
    map { $result += $_ } times;
    $result;
  }

sub print_timing($$$)
  {
    my ($message,$wall,$cpu) = @_;
    printf "%s: %.3fs wall, %.3fs cpu.\n", $message, $wall, $cpu;
  }

sub print_memory_use()
  {
    # Different operating systems have different data and ps uses different
    # arguments so try a series of commands in an attempt to collect the
    # information.
    system qw(ps -o vsz -p), $$;
    system qw(ps -l -p), $$;
    system qw(ps v), $$;
  }

FEM def para;r;femur;example
FEM def coor;r;femur;example
FEM def base;r;femur;example

#
# Reads in an initial mesh and calculates linear approximations
#  for the derivatives
#
FEM def node;r;femur;example
FEM def elem;r;femur;example
FEM up node deriv 1 linear node 2..401
FEM up node deriv 2 linear node 2..401
FEM update scale_factor normalise

#
# The initial mesh
#
FEM export node;femur_init as femur_init offset 10000
FEM export elem;femur_init as femur_init offset_elem 10000


#
# Calculate the intial xi projections
#
FEM def data;r;femur;example
FEM def xi;c close
#FEM def xi;w;femur 
FEM li data err 
FEM export data;femur_data as femur_data error


#
# Setup the fitting problem
#
FEM up field from geometry
FEM def fit;r;femur;example geometry

($wall_setup,$cpu_setup) = (time, cpu_time);
print_timing 'setup', $wall_setup - $^T, $cpu_setup;

#
# First fit
#
FEM fit
FEM update node fit
FEM update scale_factor normalise 
FEM def xi;c close old
FEM li data err 
FEM export data;femur_data_1 as femur_data_1 error
FEM export node;fitted_femur_1 as fitted_femur_1 offset 20000
FEM export elem;fitted_femur_1 as fitted_femur_1 offset_elem 20000

if( $TESTING ) {
  fem list data;fitted1 errors;
  fem list node;fitted1
}

if ($MULTIPLEFITS==1)
{

#
# Second fit
#
FEM def fit;r;femur_2;example geometry


FEM fit
FEM update node fit
FEM update scale_factor normalise
FEM def xi;c close old
FEM li data err 
FEM export data;femur_data_2 as femur_data_2 error
FEM export node;fitted_femur_2 as fitted_femur_2 offset 30000
FEM export elem;fitted_femur_2 as fitted_femur_2 offset_elem 30000

#
# Third fit
#
FEM fit
FEM update node fit
FEM update scale_factor normalise
FEM def xi;c close old
FEM li data err 
FEM export data;femur_data_3 as femur_data_3 error
FEM export node;fitted_femur_3 as fitted_femur_3 offset 40000
FEM export elem;fitted_femur_3 as fitted_femur_3 offset_elem 40000

#
# Fourth fit
#
FEM fit
FEM update node fit
FEM update scale_factor normalise
FEM def xi;c close old
FEM li data err 
FEM export data;femur_data_4 as femur_data_4 error
FEM export node;fitted_femur_4 as fitted_femur_4 offset 50000
FEM export elem;fitted_femur_4 as fitted_femur_4 offset_elem 50000
#
# Fifth fit
#
FEM fit
FEM update node fit
FEM update scale_factor normalise
FEM def xi;c close old
FEM li data err 
FEM export data;femur_data_5 as femur_data_5 error
FEM export node;fitted_femur_5 as fitted_femur_5 offset 60000
FEM export elem;fitted_femur_5 as fitted_femur_5 offset_elem 60000
#
# Sixth fit
#
FEM fit
FEM update node fit
FEM update scale_factor normalise
FEM def xi;c close old
FEM li data err 
FEM export data;femur_data_6 as femur_data_6 error
FEM export node;fitted_femur_6 as fitted_femur_6 offset 70000
FEM export elem;fitted_femur_6 as fitted_femur_6 offset_elem 70000

FEM def no;w;fitted_femur_6
FEM def elem;w;fitted_femur_6

}

($wall_solve,$cpu_solve) = (time, cpu_time);
print_timing 'solve', $wall_solve - $wall_setup, $cpu_solve - $cpu_setup;
print_memory_use;

Files used by this example are:

Name             Modified     Size

example_21d.com  29-Sep-2005  3.6k
femur.ipbase     23-Aug-2002  1.5k
femur.ipcoor     23-Aug-2002  688 
femur.ipdata     23-Aug-2002  206k
femur.ipelem     23-Aug-2002  147k
femur.ipfit      13-Apr-2007  3.8k
femur.ipnode     23-Aug-2002  393k
femur.ippara     28-Sep-2005  5.9k
femur_2.ipfit    13-Apr-2007   15k

Download the entire example:

Name                      Modified     Size

examples_2_21_21d.tar.gz  27-Aug-2010  815k

Testing status by version:

Testing status by file:

• cmiss_test.log
  

  Success	hpc_cm64_irix:	cmiss_test.log.retain.
  Success	hpc_cm_irix:	cmiss_test.log.retain.
  Success	hpc_cmo64_irix:	cmiss_test.log.retain.
  Success	hpc_cmo_irix:	cmiss_test.log.retain.
  rs6000-aix
  Success	cm:	cmiss_test.log.retain.
  Success	cm-debug:	cmiss_test.log.retain.
  Success	cm64:	cmiss_test.log.retain.
  Success	cm64-debug:	cmiss_test.log.retain.
  x86_64-linux
  Success	cm:	cmiss_test.log.retain.
  Success	cm-debug:	cmiss_test.log.retain.

  
  
• cmiss_test.out
  

  Success	hpc_cm64_irix:	ndiff test: no significant differences with generic answer.
  Success	hpc_cm_irix:	ndiff test: no significant differences with generic answer.
  Success	hpc_cmo64_irix:	ndiff test: no significant differences with generic answer.
  Success	hpc_cmo_irix:	ndiff test: no significant differences with generic answer.

  
  
• fitted1.opdata
  

  rs6000-aix
  Success	cm:	ndiff test: no significant differences with generic answer.
  Success	cm-debug:	ndiff test: no significant differences with generic answer.
  Success	cm64:	ndiff test: no significant differences with generic answer.
  Success	cm64-debug:	ndiff test: no significant differences with generic answer.
  x86_64-linux
  Success	cm:	ndiff test: no significant differences with generic answer.
  Success	cm-debug:	ndiff test: no significant differences with generic answer.

  
  
• fitted1.opnode
  

  rs6000-aix
  Success	cm:	ndiff test: no significant differences with generic answer.
  Success	cm-debug:	ndiff test: no significant differences with generic answer.
  Success	cm64:	ndiff test: no significant differences with generic answer.
  Success	cm64-debug:	ndiff test: no significant differences with generic answer.
  x86_64-linux
  Success	cm:	ndiff test: no significant differences with generic answer.
  Success	cm-debug:	ndiff test: no significant differences with generic answer.

  
  

Graphical output from this problem is given here.

Html last generated: Sun Mar 6 05:50:10 2016

Input last modified: Thu Aug 26 13:09:40 2010