18.11.2014 Views

Pit Pattern Classification in Colonoscopy using Wavelets - WaveLab

Pit Pattern Classification in Colonoscopy using Wavelets - WaveLab

Pit Pattern Classification in Colonoscopy using Wavelets - WaveLab

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

4.4 <strong>Classification</strong><br />

ν = 2 b with b l ≤ b ≤ b u and b, b l , b u ∈ Z (4.64)<br />

where a l and a u are the lower and the upper bound for parameter a and b l and b u are the<br />

lower and the upper bound for parameter b. In our implementation the parameter ν is limited<br />

by 1 [9], thus we set b u = 0.<br />

From equations (4.63) and (4.64) it is clear that <strong>in</strong> our implementation the grid search is<br />

not optimal at all s<strong>in</strong>ce it suffers from a high granularity because a ∈ Z and b ∈ Z. It is<br />

therefore quite possible that the best choices for γ and ν are not <strong>in</strong> Z but <strong>in</strong> R. For the sake<br />

of speed and simplicity however we use the naive and faster approach and simply iterate<br />

over values out of Z regardless of the possibility to miss the best possible choices for γ and<br />

ν.<br />

59

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!