An evaluation of thematic mapper simulator data for mapping forest coverComputer-aided analysis techniques applied to Thematic Mapper Simulator (TMS) data were evaluated for the purpose of mapping forest cover types. Classification results obtained using a supervised set of training statistics and various combinations of three and four channel subsets of the seven available TMS channels are compared for the L2 (Minimum Euclidean Distance), GML (Gaussian Maximum Likelihood), and SECHO (Supervised Extraction and Classification of Homogeneous Objects) classification algorithms. SECHO performed significantly better than either of the two per-point classifiers for the untransformed data. Overall classification results of the Karhunen-Loeve transformation increased for the L2 algorithm, but decreased for both the GML and SECHO algorithms.
Document ID
19840030298
Acquisition Source
Legacy CDMS
Document Type
Conference Paper
Authors
Dean, M. E. (Purdue Univ. West Lafayette, IN, United States)
Hoffer, R. M. (Purdue University West Lafayette, IN, United States)