Título:	Qualidade da classificação automática de imagens de sensoriamento remoto em trabalhos apresentados nas edições anteriores do SBSR.
Autoria:	SCHULTZ, B. LUIZ, A. J. B. SANCHES, I. D. A. FORMAGGIO, A. R.
Afiliação:	BRUNO SCHULTZ, INPE; ALFREDO JOSE BARRETO LUIZ, CNPMA; IEDA DEL'ARCO SANCHES, INPE; ANTONIO ROBERTO FORMAGGIO, INPE.
Ano de publicação:	2015
Referência:	In: SIMPÓSIO BRASILEIRO DE SENSORIAMENTO REMOTO, 17., 2015, João Pessoa. Anais... São José dos Campos: INPE, 2015. p. 2357-2364.
Conteúdo:	Abstract: A study has been carried out considering 18 years of papers published in the Simpósio Brasileiro de Sensoriamento Remoto (Brazilian Remote Sensing Symposium) related to the subject of satellite image classification. The aim of the study was to assess the degree of progress made in thematic mapping through developments in classification algorithms, different approaches (per-pixel or object-based) and methods (unsupervised and supervised). The result of 238 reported classification experiments were quantitatively analyzed through Kappa Index (KI) results. Several parameters were used to relate the experiments, as type of approach, method, number of samples and classes, used sensor system, etc. Overall, the results showed that no significant improvement was found in KI results after 18 years SBSR. The mean and KI values was found to be 0.71 and standard deviation of 0.14. Relations between KI results and number of class, type of approach and method could not be found. Thirty one percent of the experiments analyzed did not present sufficient methodological information, thus, they were excluded from the analysis.
Thesagro:	Sensoriamento remoto
NAL Thesaurus:	Remote sensing Meta-analysis
Palavras-chave:	Automatic classification SBSR Meta-análise Classificação automática
Tipo do material:	Artigo em anais e proceedings
Acesso:	openAccess
Aparece nas coleções:	Artigo em anais de congresso (CNPMA)