Title :
ART-MMAP: a neural network approach to subpixel classification
Author :
Liu, Weiguo ; Seto, Karen C. ; Wu, Elaine Y. ; Gopal, Sucharita ; Woodcock, Curtis E.
Author_Institution :
ACI Worldwide Inc., Riverside, RI, USA
Abstract :
Global or continental-scale land cover mapping with remote sensing data is limited by the spatial characteristics of satellites. Subpixel-level mapping is essential for the successful description of many land cover patterns with spatial resolution of less than ∼1 km and also useful for finer resolution data. This paper presents a novel adaptive resonance theory MAP (ARTMAP) neural network-based mixture analysis model-ART mixture MAP (ART-MMAP). Compared to the ARTMAP model, ART-MMAP has an enhanced interpolation function that decreases the effect of category proliferation in ARTa and overcomes the limitation of class category in ARTb. Results from experiments demonstrate the superiority of ART-MMAP in terms of estimating the fraction of land cover within a single pixel.
Keywords :
ART neural nets; geophysical signal processing; image classification; image resolution; interpolation; terrain mapping; vegetation mapping; ART-MMAP; adaptive resonance theory; category proliferation; class category limitation; continental-scale land cover mapping; global-scale land cover mapping; interpolation function; land cover patterns; mixture analysis; neural network; remote sensing data; satellite data; spatial characteristics; spatial resolution; subpixel classification; subpixel-level mapping; Classification tree analysis; Image resolution; Life estimation; Neural networks; Regression tree analysis; Remote sensing; Resonance; Satellite broadcasting; Spatial resolution; Subspace constraints; ART; ART-MMAP; ARTMAP; Adaptive resonance theory; adaptive resonance theory MAP; mixture MAP; mixture analysis; neural network; subpixel classification;
Journal_Title :
Geoscience and Remote Sensing, IEEE Transactions on
DOI :
10.1109/TGRS.2004.831893