source: prdatasets/emim.m @ 151

Last change on this file since 151 was 150, checked in by bduin, 5 years ago
File size: 1.1 KB
RevLine 
[80]1%EMIM Load 8-band electron microscope images
2%
3%       A = EMIM(N)
4%
5% Return image N as the dataset A. In this dataset the pixels of an
6% electron-microscope image are represented as objects with 8 features.
7% These represent the bands of the EM image. There are 5 images available,
8% encoded by N = 31, 32, 33, 34 and 37. Default is N = 31. The first four
9% images have size 128x128. The image with N=37 has size 256x256.
10%
11% The segmentation of such images has been studied in:
12% P. Paclik, R.P.W. Duin, G.M.P. van Kempen, and R. Kohlus, Segmentation of
13% multi-spectral images using the combined classifier approach, Image and
14% Vision Computing Journal, vol. 21, no. 6, 2003, 473-482.
15%
16% See also DATASETS, PRDATASETS
17
[150]18% Copyright: R.P.W. Duin
[80]19% Faculty EWI, Delft University of Technology
20% P.O. Box 5038, 2600 GA Delft, The Netherlands
21
22function a = emim(n)
23
24if nargin ==0, n=31; end
25
26if all([31,32,33,34,37]~=n)
27        error('Images should be coded by N = 31, 32, 33, 34 or 37')
28end
29
[137]30name = ['emim' num2str(n) '.mat'];
31a = pr_getdata(['http://prtools.tudelft.nl/prdatasets/' name],[],name);
[80]32a = setname(a,['EM' num2str(n) ' Image Delft']);
33
Note: See TracBrowser for help on using the repository browser.