deletions | additions       diff --git a/section_Texton_based_Machine_Learning__.tex b/section_Texton_based_Machine_Learning__.tex  index 7eeff7d..755bd63 100644  --- a/section_Texton_based_Machine_Learning__.tex  +++ b/section_Texton_based_Machine_Learning__.tex 
...
similar histograms.  \citet{de2009design} use textons as image features for distinguishing  between three height classes during flight~\cite{de2009design}. flight.  Using a nearest neighbor classifier, their approach achieves a height  classification error of approximately 22\,\% on a hold-out test set.  This enables a flapping-wing MAV during an experiment to roughly hold  its height. In another work, \citet{de2012appearance} introduce  the \emph{appearance variation cue} cue}, which is based on textons,  for estimating the proximity to objects~\cite{de2012appearance}. Since closer objects should have less  variation, these objects should appear less varied. Using this method,  their MAV is successfully able to avoid obstacles in a $5m \times 5m$