Experiments 24112004-16-24

Checking different texture OF's

///////////////////////////////////////////////////////////

// 24/11/2004: Trying to see how texture OF's behave

// Comparison with various other methods

// 20 intensive iterations of NRR, two levels

// of affine.

The 9 OF's are:

1. 24112004-16: zero
2. 24112004-17: ssd
3. 24112004-18: dot
4. 24112004-19: abs_diff
5. 24112004-20: wtd_abs_diff
6. 24112004-21: eigen
7. 24112004-22: mi
8. 24112004-23: nmi
9. 24112004-24: res_info

Conclusion: See individual results in the directory which collects RESULTS from all; Quality depends on type of analysis. However, the good results are reached by using:

• ssd
• abs_diff
• wtd_abs_diff

Some methods go completely in the wrong direction, e.g. res_info, eigen and zero.

For all results in brief See: INTENSITY_OFs--24112004 in experiments archive where all final results have been put.