this is for holding javascript data
Ardo Illaste edited res_cluster.md
about 10 years ago
Commit id: e918b8a93b2e6fa65e042dbfa5f9b0b2e2076358
deletions | additions
diff --git a/res_cluster.md b/res_cluster.md
index 76bac7b..e44b976 100644
--- a/res_cluster.md
+++ b/res_cluster.md
...
### Clustering
Having determined the events in each pixel it is possible to reconstruct the image with reduced noise levels using the matrix \(E\). However, this will not tell us anything about the properties of
actual release events
(number, dynamics) (e.g., spark/wave numbers or properties) as these
macroscopic events are made up of several events from different pixels.
Therefore it It is
therefore necessary to combine
elementary events from various pixels into macroscopic release events.
This is achieved using the clustering method DBSCAN \cite{Sander_1998}. The
method works in the parameter space and finds
a clusters of arbitrary shape based on the density of
events in parameter space. events. This is preferable to standard clustering methods which often yield radially symmetric clusters (k-means, etc).
Clustering is performed twice. First pixel events are clustered accoring to their shape i.e., clustering is done on matrix \(E_s\). This step distributes pixels into several groups based on solely their
shape. For example, into shape (e.g., groups of elementary events composing spark and wave
events. events). In the second clustering step,
the \(E^p\) matrix is cluster for each shape group
is clustered based on location. This way, and physically nearby clusters of similar events are obtained. With this two-step approach, release events
of various types consisting on
elementary events from multiple pixels are obtained.