Comparison between co-occurrence and wavelet features for characterization of urban environments by SAR data

It is well known that co-occurrence matrix related measures have already shown to be effective in remote sensing. While wavelet analysis, even if widely applied to texture classification problems, has not been evaluated in the context of urban environment classification. In this paper we present a comparative study between these two textural features for characterization of urban environments by SAR data. We found that the correct classification rates are slightly superior by using wavelet features.