%0 Conference Proceedings
%@holdercode {isadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S}
%@nexthigherunit 8JMKD3MGPCW/3ER446E
%@usergroup administrator
%@usergroup simone
%3 landsat data processing.pdf
%B Geoscience and Remote Sensing Symposium, IGARSS'89; Canadian Symposium on Remote Sensing, 12.
%X This paper presents results from an experiment performed to assess the performance of the median recurrent filter for stripping removal from TM/LANDSAT data. The experiment consisted of comparing the resulting images in terms of their potential for water and land use classification. This comparison was perfomed taking  into account two types of remote sensor data user: the user interested in manual interpretation. As far as the potential for manual interpretation is concerned, the different images were compared in terms of the digital filtering effects on the elements of importance for the information extraction: color, texture, shape, contrast, etc. The image performance in terms of digital classification was assessed by comparing the classification matrix resulting from the application of a maximum likelihood algorithm to each data set: Reflectance image; reflectance filtering image; Composite image. The performance of each data set was assessed through the percent of correct classification produced by the samples used to train the system. The results indicate that the filtered image allowed greater separability between opticallydistinct water classes, presenting however a poor performance for non-aquatic targets as far as digital analysis was concerned. For manual analysis, however, the composite image was the best.
%@mirrorrepository sid.inpe.br/mtc-m18@80/2008/03.17.15.17.24
%T TM/Landsat data processing for inland water monitoring
%@electronicmailaddress
%@electronicmailaddress evlyn@ltid.inpe.br
%@secondarytype PRE CI
%8 10-14 July
%@group
%@group DSR-INPE-BR
%@secondarykey INPE--PRE/
%2 sid.inpe.br/mtc-m18@80/2009/08.19.19.44.43
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%4 sid.inpe.br/mtc-m18@80/2009/08.19.19.44
%D 1989
%S Proceedings
%A Godoy Júnior, Moacir,
%A Novo, Evlyn Márcia Leão de Moraes,
%C Vancouver, CA
%@area SRE