ART-MMAP: A neural network approach to subpixel classification

被引：44
作者：
Liu, WG [1 ]
Seto, KC
Wu, EY
Gopal, S
Woodcock, CE
机构：
[1] ACI Worldwide Inc, Riverside, RI 02915 USA
[2] Stanford Univ, Dept Geol & Environm Sci, Stanford, CA 94305 USA
[3] Med Univ S Carolina, Dept Ophthalmol, Charleston, SC 29425 USA
[4] Boston Univ, Dept Geog, Boston, MA 02215 USA
来源：
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING | 2004年 / 42卷 / 09期
基金：
美国国家科学基金会; 美国国家航空航天局;
关键词：
adaptive resonance theory (ART) mixture MAP (ART-MMAP); adaptive resonance theory MAP (ARTMAP); mixture analysis; neural network; subpixel classification;
D O I：
10.1109/TGRS.2004.831893
中图分类号：
P3 [地球物理学]; P59 [地球化学];
学科分类号：
0708 ; 070902 ;
摘要：
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 similar to1 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 ART, and overcomes the limitation of class category in ART(b). Results from experiments demonstrate the superiority of ART-MMAP in terms of estimating the fraction of land cover within a single pixel.
引用
收藏
页码：1976 / 1983
页数：8
相关论文
共 31 条
[1]   CLASSIFICATION OF MULTISPECTRAL IMAGES BASED ON FRACTIONS OF ENDMEMBERS - APPLICATION TO LAND-COVER CHANGE IN THE BRAZILIAN AMAZON [J].
ADAMS, JB ;
SABOL, DE ;
KAPOS, V ;
ALMEIDA, R ;
ROBERTS, DA ;
SMITH, MO ;
GILLESPIE, AR .
REMOTE SENSING OF ENVIRONMENT, 1995, 52 (02) :137-154
[2]   On estimating measurement error in remotely-sensed images with the variogram [J].
Atkinson, PM .
INTERNATIONAL JOURNAL OF REMOTE SENSING, 1997, 18 (14) :3075-3084
[3]   Constructive feedforward ART clustering networks - Part I [J].
Baraldi, A ;
Alpaydin, E .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2002, 13 (03) :645-661
[4]   NONLINEAR SPECTRAL MIXING MODELS FOR VEGETATIVE AND SOIL SURFACES [J].
BOREL, CC ;
GERSTL, SAW .
REMOTE SENSING OF ENVIRONMENT, 1994, 47 (03) :403-416
[5]  
Carpenter G., 1991, Pattern recognition by self-organizing neural networks
[6]   Distributed ARTMAP: a neural network for fast distributed supervised learning [J].
Carpenter, GA ;
Milenova, BL ;
Noeske, BW .
NEURAL NETWORKS, 1998, 11 (05) :793-813
[7]   FUZZY ART - FAST STABLE LEARNING AND CATEGORIZATION OF ANALOG PATTERNS BY AN ADAPTIVE RESONANCE SYSTEM [J].
CARPENTER, GA ;
GROSSBERG, S ;
ROSEN, DB .
NEURAL NETWORKS, 1991, 4 (06) :759-771
[8]   A neural network method for mixture estimation for vegetation mapping [J].
Carpenter, GA ;
Gopal, S ;
Macomber, S ;
Martens, S ;
Woodcock, CE .
REMOTE SENSING OF ENVIRONMENT, 1999, 70 (02) :138-152
[9]   FUZZY ARTMAP - A NEURAL NETWORK ARCHITECTURE FOR INCREMENTAL SUPERVISED LEARNING OF ANALOG MULTIDIMENSIONAL MAPS [J].
CARPENTER, GA ;
GROSSBERG, S ;
MARKUZON, N ;
REYNOLDS, JH ;
ROSEN, DB .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1992, 3 (05) :698-713
[10]   ARTMAP - SUPERVISED REAL-TIME LEARNING AND CLASSIFICATION OF NONSTATIONARY DATA BY A SELF-ORGANIZING NEURAL NETWORK [J].
CARPENTER, GA ;
GROSSBERG, S ;
REYNOLDS, JH .
NEURAL NETWORKS, 1991, 4 (05) :565-588
1234