Home > Source > Evaluation > eval_groupwise_multi_point.m

eval_groupwise_multi_point

PURPOSE ^

OPTIMISE_GROUPWISE: Uses PDF's to minimise group-wise complexity

SYNOPSIS ^

function [score, warped_points] = eval_groupwise_multi_point(knot_displacements, init_knots, unwarped_image_vec, unwarped_points_vec, image_set, points_set, n_modes, start_points_vec, warp_params, model_evaluation_method, weighting_normalisation_method, shape_weight, variation_kept, pdf_evaluation_method, kernel)

DESCRIPTION ^

 OPTIMISE_GROUPWISE: Uses PDF's to minimise group-wise complexity

    GENERAL

      function [score, warped_points] = 
       eval_groupwise_multi_point(knot_displacements, init_knots,
       unwarped_image_vec, unwarped_points_vec, image_set, points_set,
       n_modes, start_points_vec, warp_params, model_evaluation_method,
       weighting_normalisation_method, shape_weight, variation_kept)

 
    INPUT/S

      -X:
          X
           
    OUTPUT/S

      -X:
           X

    PENDING WORK
      
      

    KNOWN BUG/S

      

    COMMENT/S

      

    RELATED FUNCTION/S

      

    ABOUT

      -Created:     February 9th, 2004
      -Last update: February 2004
      -Revision:    0.1.0
      -Author:      R. S. Schestowitz, University of Manchester
 ==============================================================

CROSS-REFERENCE INFORMATION ^

This function calls: This function is called by:

SOURCE CODE ^

0001 function [score, warped_points] = eval_groupwise_multi_point(knot_displacements, init_knots, unwarped_image_vec, unwarped_points_vec, image_set, points_set, n_modes, start_points_vec, warp_params, model_evaluation_method, weighting_normalisation_method, shape_weight, variation_kept, pdf_evaluation_method, kernel)
0002 % OPTIMISE_GROUPWISE: Uses PDF's to minimise group-wise complexity
0003 %
0004 %    GENERAL
0005 %
0006 %      function [score, warped_points] =
0007 %       eval_groupwise_multi_point(knot_displacements, init_knots,
0008 %       unwarped_image_vec, unwarped_points_vec, image_set, points_set,
0009 %       n_modes, start_points_vec, warp_params, model_evaluation_method,
0010 %       weighting_normalisation_method, shape_weight, variation_kept)
0011 %
0012 %
0013 %    INPUT/S
0014 %
0015 %      -X:
0016 %          X
0017 %
0018 %    OUTPUT/S
0019 %
0020 %      -X:
0021 %           X
0022 %
0023 %    PENDING WORK
0024 %
0025 %
0026 %
0027 %    KNOWN BUG/S
0028 %
0029 %
0030 %
0031 %    COMMENT/S
0032 %
0033 %
0034 %
0035 %    RELATED FUNCTION/S
0036 %
0037 %
0038 %
0039 %    ABOUT
0040 %
0041 %      -Created:     February 9th, 2004
0042 %      -Last update: February 2004
0043 %      -Revision:    0.1.0
0044 %      -Author:      R. S. Schestowitz, University of Manchester
0045 % ==============================================================
0046 
0047 % scale it all down to try and get it to move more!
0048 % init_knots = init_knots*0.01;
0049 % start_points_vec = start_points_vec*0.01;
0050 
0051 warped_points = nrr_trans_1d(start_points_vec, init_knots, init_knots + knot_displacements, warp_params, []);
0052 warped_image = interp1(unwarped_points_vec,unwarped_image_vec,warped_points, 'linear',0);
0053 image_set(:,size(image_set,2)+1) = warped_image;
0054 points_set(:,size(points_set,2)+1) = warped_points;
0055 
0056 groupwise_pdf = compute_pdf(image_set, points_set, kernel, shape_weight);
0057 score = measure_pdf(groupwise_pdf, pdf_evaluation_method, warped_image);
0058 
0059 % score = 0; % to be modified w.r.t. eval_groupwise_single_point

Generated on Fri 14-May-2004 10:05:30 by m2html © 2003