An assessment of SPOT XS and Landsat MSS data for digital classification of near-urban land cover

Abstract

The potential of SPOT Multispectral (XS) and Landsat Multispectral Scanner (MSS) for automatically outlining agricultural near-urban interfaces, by means of classifications, was tested for a study area around Yogyakarta, Central Java. Photo-interpretation of SPOT panchromatic (P) images and aerial photographs acquired during the past 20 years was used as a reference to evaluate maps derived from SPOT XS and Landsat MSS and to quantify the evolution of this area. SPOT XS mapping proved to be reliable, whereas only overall information could be derived from Landsat. Due to the usual heterogeneity of the Javanese landscapes and to the present unavailability of locally-adapted lextural classifiers, the visual analysis of SPOT P images provided better results than the digital processing of SPOT XS data.