Martin Coath edited section_Methods_For_results_derived__.tex  about 8 years ago

Commit id: 61f87bdae6afc4da6e51f1dfe375c06280cb7cc2

deletions | additions      

       

\item the number of pixels $m$ corresponding to half the required window size $w$ is calculated: $m = \mathrm{floor}(\frac{w}{2})$  \item thus the working window size is $n = 2 \cdot m+1$   \item assemble the vector $\vec{k}_{j}$ of values for a horizontal window of $n$ contiguous pixels $k_{[i,j-m \: : \: i,j \: : \: i,j+m]}$  \item repeat with $\vec{k}_{i}$, a vector of values for a vertical window of $n$ pixels $k_{[i-m,j \:  : \:  i,j \:  : \:  i+m,j]}$ \item both vectors are normalized, so they can be treated as distributions, and the Skewness $\gamma$ of each distribution, \textit{i.e.} the asymmetry in the gray-scale values in both direction, is calculated, $\gamma_j$ and $\gamma_i$  \item the \textsc{skv} value of the pixel is the larger of the two values $\max(\gamma_j, \gamma_i)$  \end{enumerate}