MAKE_1D_MODEL: Genrates examples of 1-D images. Code written by Katherine Smith, 2003 GENERAL [imagelist,images,points, his, los] = make_1d_images(n_images, image_width, white_width) INPUT/S -n_images: The number of images to generate. -image_width: The width of the 1-D images to be created. -white_width: The width of the white component?? OUTPUT/S -imagelist: The list of the images generated. -images: The images generated. -points: The (control) points of the images generted. -his: The upper bound of the examples generated? IT APPEARS TO BE NUMBER OF POINTS WHERE BUMP IS HIGH -los: The lower bound of the examples generated? PENDING WORK -Understand function better. -Improve input/output documentation. KNOWN BUG/S - COMMENT/S -Some comments added 28/11/03. The notion of 'white' is still not understood. -Most comments are old and should be ignored (2004) RELATED FUNCTION/S ABOUT -Created: November 23rd, 2003 -Last update: Novermber 28th, 2003 -Revision: 0.0.4 -Author: R. S. Schestowitz, University of Manchester ==============================================================
0001 function [imagelist,images,points, his, los] = make_1d_images(n_images, image_width, white_width) 0002 % MAKE_1D_MODEL: Genrates examples of 1-D images. 0003 % 0004 % Code written by Katherine Smith, 2003 0005 % 0006 % GENERAL 0007 % 0008 % [imagelist,images,points, his, los] = 0009 % make_1d_images(n_images, image_width, white_width) 0010 % 0011 % INPUT/S 0012 % 0013 % -n_images: 0014 % The number of images to generate. 0015 % 0016 % -image_width: 0017 % The width of the 1-D images to be created. 0018 % 0019 % -white_width: 0020 % The width of the white component?? 0021 % 0022 % OUTPUT/S 0023 % 0024 % -imagelist: 0025 % The list of the images generated. 0026 % 0027 % -images: 0028 % The images generated. 0029 % 0030 % -points: 0031 % The (control) points of the images generted. 0032 % 0033 % -his: 0034 % The upper bound of the examples generated? IT 0035 % APPEARS TO BE NUMBER OF POINTS WHERE BUMP IS HIGH 0036 % 0037 % -los: 0038 % The lower bound of the examples generated? 0039 % 0040 % PENDING WORK 0041 % 0042 % -Understand function better. 0043 % -Improve input/output documentation. 0044 % 0045 % KNOWN BUG/S 0046 % 0047 % - 0048 % 0049 % COMMENT/S 0050 % 0051 % -Some comments added 28/11/03. The notion of 'white' is 0052 % still not understood. 0053 % -Most comments are old and should be ignored (2004) 0054 % 0055 % RELATED FUNCTION/S 0056 % 0057 % 0058 % 0059 % ABOUT 0060 % 0061 % -Created: November 23rd, 2003 0062 % -Last update: Novermber 28th, 2003 0063 % -Revision: 0.0.4 0064 % -Author: R. S. Schestowitz, University of Manchester 0065 % ============================================================== 0066 0067 unwarped_points = (1:image_width)'; 0068 % set original unwarped points to be diagonal 1..50 0069 % unwarped points is then a column vector [1;2;3...] 0070 % size of <unwarped_points> is 50,1 0071 images = zeros([image_width n_images]); 0072 % set a bunch of black images of appropriate 0073 % size (rather a flat horizontal line at y=0) 0074 points = unwarped_points(:,ones(n_images,1)); 0075 % set points to be a copy of 0076 % unwarped points but of width ten 0077 for i=1:n_images 0078 % for all images 0079 % image = zeros(image_width,1); 0080 white = 0.75 + (rand-0.5)*0.2; 0081 % get some bright shade of white (gray) 0082 % range will be .75 to 0.85 0083 % white = 0.9; 0084 this_white_width = rand*(image_width/5) + white_width*image_width; 0085 % get random up to 20% of image width and add to 50 times 0086 % white_width which is input often assigned to 0.2 0087 % so values of this_white_width lie between 10 and 30 0088 if(this_white_width > 0) 0089 % if this white shade/value exists 0090 los(i) = floor((image_width+1)/2-this_white_width/2); 0091 % set high and low value 0092 his(i) = ceil((image_width+1)/2+this_white_width/2); 0093 % Is this the half bumb which is symmetric?? 0094 for j=los(i):his(i) 0095 % within this range... 0096 % image(j) = white; 0097 images(j,i) = white; 0098 end 0099 end 0100 filename = ['1dim' num2str(i) '.png']; 0101 % set name of file 0102 % imwrite(images(i,:), filename, 'png'); 0103 % write file 0104 imagelist{i} = filename; 0105 % save in list of images 0106 %figure, plot(image); 0107 his = his'; 0108 % returns the highs and lows in correct form 0109 los = los'; 0110 end