Sub-pixel classification of MODIS images

  • Authors:

  • Hasan Roosta;Mohammad Reza Saradjian

  • Affiliations:

  • Center of Excellence in Geomatic Eng. and Disaster Management, Dept. of Surveying and Geomatic Eng., Faculty of Eng., University of Tehran, Tehran, Iran;Center of Excellence in Geomatic Eng. and Disaster Management, Dept. of Surveying and Geomatic Eng., Faculty of Eng., University of Tehran, Tehran, Iran

  • Venue:
  • NOLASC'07 Proceedings of the 6th WSEAS international conference on Non-linear analysis, non-linear systems and chaos
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

The Moderate Resolution Imaging Spectroradiometer (MODIS) offers a unique combination of spectral, temporal, and spatial resolution compared to previous global sensors, making it a good candidate for large-scale crop type mapping. However, because of sub-pixel heterogeneity, the application of traditional hard classification approach to MODIS may result in significant errors in crop area estimation. This study examined the potential of sub-pixel classification for regional crop area estimation using time series of NDVI-composites of MODIS. Fars province in south of Iran was selected as test zone, because of the cover type of the large majority of agricultural fields. Neural network model was investigated and its result in area fraction images (AFIs). The AFIs contain for each 250 m pixel the estimated area proportions occupied by the different cover types (crops or other land use). The algorithm was trained with both of reference data and in situ data which collected by GPS in Marvdasht District. For the major classes (winter wheat, maize and other crops) the obtained acreage estimates showed good agreement with the true values (R2≅ 90%). The method seems attractive for wide-scale, regional area estimation in the countries that appropriate data are not available.