Bayesian and geostatistical approaches to combining categorical data derived from visual and digital processing of remotely sensed images

This paper seeks a synthesis of Bayesian and geostatistical approaches to combining categorical data in the context of remote sensing classification. By experiment with aerial photographs and Landsat TM data, accuracy of spectral, spatial, and combined classification results was evaluated. It was confirmed that the incorporation of spatial information in spectral classification increases accuracy significantly. Secondly, through test with a 5-class and a 3-class classification schemes, it was revealed that setting a proper semantic framework for classification is fundamental to any endeavors of categorical mapping and the most important factor affecting accuracy. Lastly, this paper promotes non-parametric methods for both definition of class membership profiling based on band-specific histograms of image intensities and derivation of spatial probability via indicator kriging, a non-parametric geostatistical technique.

[1]  Daniel L. Civco,et al.  Evidential Reasoning-Based Classification of Multi-Source Spatial Data for Improved Land Cover Mapping , 1994 .

[2]  Phaedon C. Kyriakidis,et al.  Geostatistical Mapping of Thematic Classification Uncertainty , 2004 .

[3]  Pierre Goovaerts,et al.  Geostatistical incorporation of spatial coordinates into supervised classification of hyperspectral data , 2002, J. Geogr. Syst..

[4]  Manfred Ehlers,et al.  Multisensor image fusion techniques in remote sensing , 1991 .

[5]  Alan H. Strahler,et al.  The Use of Prior Probabilities in Maximum Likelihood Classification , 1980 .

[6]  A. Skidmore.  An expert system classifies eucalypt forest types using thematic mapper data and a digital terrain model , 1989 .

[7]  P. Mather,et al.  Classification Methods for Remotely Sensed Data , 2001 .

[8]  F. E. Townsend.  The enhancement of computer classifications by logical smoothing , 1986 .

[9]  John A. Richards,et al.  Knowledge-based techniques for multi-source classification , 1990 .

[10]  D. Peddle,et al.  Image texture processing and data integration for surface pattern discrimination , 1991 .

[11]  R. Tibshirani,et al.  Combining Estimates in Regression and Classification , 1996 .

[12]  Michel M. Verstraete,et al.  Potential and limitations of information extraction on the terrestrial biosphere from satellite remote sensing , 1996 .

[13]  Fabio Maselli,et al.  Use of error matrices to improve area estimates with maximum likelihood classification procedures , 1992 .

[14]  Brian M. Steele,et al.  Combining Multiple Classifiers: An Application Using Spatial and Remotely Sensed Information for Land Cover Type Mapping , 2000 .

[15]  Fabio Maselli,et al.  Integration of ancillary data into a maximum-likelihood classifier with nonparametric priors , 1995 .

[16]  Michael D. Steven.  Ground truth An underview , 1987 .

[17]  Carolyn T. Hunsaker,et al.  Spatial uncertainty in ecology : implications for remote sensing and GIS applications , 2002 .

[18]  Charlotte M. Gurney,et al.  The use of contextual information to improve land cover classification of digital remotely sensed data , 1981 .

[19]  M. Goodchild,et al.  Uncertainty in geographical information , 2002 .

[20]  Christopher O. Justice,et al.  Information extraction from remotely sensed data. , 1981 .

[21]  A. K. Bregt,et al.  Comparison of thematic maps derived from a soil map and from kriging of point data. , 1987 .

[22]  Giles M. Foody,et al.  Fully-fuzzy supervised classification of sub-urban land cover from remotely sensed imagery: Statistical and artificial neural network approaches , 2001 .

[23]  Phaedon C. Kyriakidis,et al.  Geostatistical Models of Uncertainty for Spatial Data , 2001 .

[24]  Terry L. Sohl,et al.  Regional characterization of land cover using multiple sources of data , 1998 .

[25]  John Aitchison,et al.  The statistical analysis of geochemical compositions , 1984 .

[26]  Tong Lee,et al.  Probabilistic and Evidential Approaches for Multisource Data Analysis , 1987, IEEE Transactions on Geoscience and Remote Sensing.

[27]  S. Bruin.  Predicting the areal extent of land-cover types using classified imagery and geostatistics. , 2000 .

[28]  G. Lugosi,et al.  Consistency of Data-driven Histogram Methods for Density Estimation and Classification , 1996 .

[29]  J. Townshend,et al.  A new global 1‐km dataset of percentage tree cover derived from remote sensing , 2000 .

[30]  D. H. Card.  Using known map category marginal frequencies to improve estimates of thematic map accuracy , 1982 .

[31]  Stuart Barr,et al.  Reducing structural clutter in land cover classifications of high spatial resolution remotely-sensed images for urban land use mapping , 2000 .

[32]  Johannes R. Sveinsson,et al.  Hybrid consensus theoretic classification , 1996, IGARSS '96. 1996 International Geoscience and Remote Sensing Symposium.

[33]  Ronald J. P. Lyon,et al.  A prior probability method for smoothing discriminant analysis classification maps , 1982 .

[34]  John C. Davis,et al.  Estimation of regionalized compositions: A comparison of three methods , 1995 .

[35]  Russell G. Congalton,et al.  A review of assessing the accuracy of classifications of remotely sensed data , 1991 .

[36]  Andre G. Journel,et al.  Joint simulation of multiple variables with a Markov-type coregionalization model , 1994 .

[37]  S. Stehman,et al.  Statistical sampling to characterize recent United States land-cover change , 2003 .

[38]  Stephen V. Stehman,et al.  Design and Analysis for Thematic Map Accuracy Assessment: Fundamental Principles , 1998 .