this is for holding javascript data
Lucas Fidon edited subsection_Approximation_of_probability_distribution__.tex
almost 8 years ago
Commit id: b5ea2ebfddd12bf9fa122882e8c753ab9c24eb6f
deletions | additions
diff --git a/subsection_Approximation_of_probability_distribution__.tex b/subsection_Approximation_of_probability_distribution__.tex
index 9b4d9e7..4d4170d 100644
--- a/subsection_Approximation_of_probability_distribution__.tex
+++ b/subsection_Approximation_of_probability_distribution__.tex
...
Hence the following definition of the probability of $(x,y)$ given $T$:
\[P((x,y),T) = \sum_{(x',y')\in T}K((x,y),(x',y'))\]
where $K$ is a gaussian kernel. in practice we take a discrete gaussian kernel filter for $K$.
\begin{table}
\begin{tabular}{ c | c | c } \[ K = \left| \begin{array}{ccc}
0.0625 & 0.125 & 0.0625 \\
0.125 & 0.25 & 0.125 \\
0.0625 & 0.125 & 0.0625
\\
\end{tabular}
\end{table} \end{array} \right|.\]
Whereas the simple histogram method places a spike function (i.e. $K = \delta$) at the bin corresponding to $(x,y)$ and update only a single bin, Parzen windowing places a kernel at the bin of $(x,y)$ and updates all bins falling under the kernel with the corresponding kernel value.
As a result using a gaussian filter, the estimated distributions are more smooth and less sparse.