Focal-Test-Based Spatial Decision Tree Learning: A Summary of Results
Authors:
- Zhe JiangZJZhe Jiang
- Shashi ShekharSSShashi Shekhar
- Xun ZhouXZXun Zhou
- Joseph KnightJKJoseph Knight
- Jennifer CorcoranJCJennifer Corcoran
- 1 December 2013
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15504786,p. 320-329
Abstract
Given a raster spatial framework, as well as training and test sets, the spatial decision tree learning (SDTL) problem aims to minimize classification errors as well as salt-and-pepper noise. The SDTL problem is important due to many societal applications such as land cover classification in remote sensing. However, the SDTL problem is challenging due to the spatial autocorrelation of class labels, and the potentially exponential number of candidate trees. Related work is limited due to the use of local-test-based decision nodes, which can not adequately model spatial autocorrelation during test phase, leading to high salt-and-pepper noise. In contrast, we propose a focal-test-based spatial decision tree (FTSDT) model, where the tree traversal direction for a location is based on not only local but also focal (i.e., neighborhood) properties of the location. Experimental results on real world remote sensing datasets show that the proposed approach reduces salt-and-pepper noise and improves classification accuracy.Keywords
This publication has 25 references indexed in Scilit:
- Global and Local Spatial Autocorrelation in Predictive Clustering Trees2011
- Spectral–Spatial Classification of Hyperspectral Imagery Based on Partitional Clustering TechniquesIEEE Transactions on Geoscience and Remote Sensing, 2009
- Spatial defect pattern recognition on semiconductor wafers using model-based clustering and Bayesian inferenceEuropean Journal of Operational Research, 2008
- Geographic Object-Based Image Analysis (GEOBIA): A new name for a new discipline2008
- A Spatial Entropy-Based Decision Tree for Classification of Geographical InformationTransactions in GIS, 2006
- The utility of texture analysis to improve per‐pixel classification for high to very high spatial resolution imageryInternational Journal of Remote Sensing, 2005
- Ensembles of Classifiers for Morphological Galaxy ClassificationThe Astrophysical Journal, 2001
- A Markov random field model for classification of multisource satellite imageryIEEE Transactions on Geoscience and Remote Sensing, 1996
- A survey of decision tree classifier methodologyIEEE Transactions on Systems, Man, and Cybernetics, 1991
- Induction of decision treesMachine Learning, 1986