infotech

 

 

Feedback and inquiries

Center for Machine Vision Research
Department of Computer Science and Engineering P.O.Box 4500
FIN-90014 University of Oulu
Finland
Tel. +358-8-553 2525
contact-cse (at) ee.oulu.fi

University of Oulu
Outex Texture Database

Contributed segmentation test suites

There is one test suite in this section.


Test suite ID: Contrib_SS_00000
Segmentation type: Supervised
Number of problems: 12
Description of the problem: The experiment involves images from three different sources: the Brodatz album, the MIT Vision Texture database, and the MeasTex database. Consequently, images captured with different equipment and under different conditions are used. For each texture present in a test mosaic there is a 256x256 training image that is extracted from a different area in the source image so that an unbiased error estimate is obtained. Since the source images were globally histogram equalized prior to being used, the gray level mean and deviation of a training image, and the corresponding texture in the test mosaic are roughly equal.
Publication: Texture Classification by Multi-Predicate Local Binary Pattern Operators
Download

 

 

Contributed classification test suites

There are 7 test suites in this section.
Use the links below to see detailed descriptions.


Test suite ID: Contrib_TC_00000
Illumination invariance: No
Rotation invariance: Yes
Spatial scale invariance: No
Color textures: No
Cost functions: No
Different class
frequencies:
No
Number of problems: 1
Description of the problem: Image data included 16 texture classes from the Brodatz album. For each texture class there were eight 256x256 images, of which the first was used for training the classifier, while the other seven images were used to test the classifier.
Publication: Gray scale and rotation invariant texture classification with Local Binary Patterns
Download  


Test suite ID: Contrib_TC_00001
Illumination invariance: No
Rotation invariance: Yes
Spatial scale invariance: No
Color textures: No
Cost functions: No
Different class
frequencies:
No
Number of problems: 10
Description of the problem: Image data included 16 texture classes from the Brodatz album. For each texture class there were eight 256x256 images. Classifier was trained with samples of just one rotation angle and tested with samples of other nine rotation angles.
Publication: Gray scale and rotation invariant texture classification with Local Binary Patterns
Download  


Test suite ID: Contrib_TC_00002
Illumination invariance: No
Rotation invariance: Yes
Spatial scale invariance: No
Color textures: No
Cost functions: No
Different class
frequencies:
No
Number of problems: 7
Description of the problem: Texture classification problem with Brodatz textures. (32 x 32 subimages)
Publication: Rotation-invariant texture classification using feature distributions
Download  


Test suite ID: Contrib_TC_00003
Illumination invariance: No
Rotation invariance: Yes
Spatial scale invariance: No
Color textures: No
Cost functions: No
Different class
frequencies:
No
Number of problems: 7
Description of the problem: Texture classification problem with Brodatz textures. (64 x 64 subimages)
Publication: Rotation-invariant texture classification using feature distributions
Download  


Test suite ID: Contrib_TC_00004
Illumination invariance: No
Rotation invariance: Yes
Spatial scale invariance: Yes
Color textures: No
Cost functions: No
Different class
frequencies:
No
Number of problems: 10
Description of the problem: Texture classification problem. 32 textures from the Brodatz album.
Publication: Texture discrimination with multidimensional distributions of signed differences
Download  


Test suite ID: Contrib_TC_00005
Illumination invariance: No
Rotation invariance: No
Spatial scale invariance: No
Color textures: No
Cost functions: No
Different class
frequencies:
No
Number of problems: 100
Description of the problem: Texture classification problem involving 11 different mixtures of barley and rice.
Publication: Determining composition of grain mixtures by texture classification based on feature distributions
Download  

 


Test suite ID: Contrib_TC_00006
Illumination invariance: No
Rotation invariance: No
Spatial scale invariance: No
Color textures: Yes
Cost functions: No
Different class
frequencies:
No
Number of problems: 1
Description of the problem: Color texture problem with VisTex images
Publication:  
Download

 

 

Last modified: 2007-10-01