Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Next revision		 Previous revision
local_binary_patterns [2009-04-21 15:05] – created davegriffithslocal_binary_patterns [2009-04-21 16:18] (current) – davegriffiths
Line 1: Line 1:
 An attempt to encode images in a way which preserves texture information and minimises lighting and other pesky variations. An attempt to encode images in a way which preserves texture information and minimises lighting and other pesky variations.
 +
 +Image data:
  
 | 5 | 4 | 3 | | 5 | 4 | 3 |
 | 4 | 3 | 1 | | 4 | 3 | 1 |
 | 2 | 0 | 3 | | 2 | 0 | 3 |
 +
 +Threshold with respect to the centre pixel:
  
 | 1 | 1 | 1 | | 1 | 1 | 1 |
 | 1 |   | 0 | | 1 |   | 0 |
 | 0 | 0 | 1 | | 0 | 0 | 1 |
 +
 +Binary encode with:
  
 | 1 | 2 | 4 | | 1 | 2 | 4 |
 | 8 |   | 16 | | 8 |   | 16 |
 | 32 | 64 | 128 | | 32 | 64 | 128 |
 +
 +To produce:
 +
 +| 1 | 2 | 4 |
 +| 8 |   | 0 |
 +| 0 | 0 | 128 |
 +
 += 143
 +
 +This number represents the relationship of the centre pixel with it's neighbours in a way which ignores overall changes in contrast or level. 
 +
 +Obviously it's not too instructive to look at these images, but obviously we are going to try:
 +
 +{{:lbp.png|}} {{:lbp2.png|}}
 +
 +Further stuff:
 +
 +  * For face detection/recognition: [[http://www.jdl.ac.cn/doc/2005/Local%20Gabor%20Binary%20Pattern%20Histogram%20Sequence%20(LGBPHS)%20A%20Novel%20Non-Statistical%20Model%20for%20Face%20Representation%20and%20Recognition.pdf]]
  
  
  • local_binary_patterns.1240326307.txt.gz
  • Last modified: 2009-04-21 15:05
  • by davegriffiths