论文信息 - A contextual classification method for recognizing land use patterns in high resolution remotely sensed data

A contextual classification method for recognizing land use patterns in high resolution remotely sensed data

Abstract This paper describes a CONtextual ANalysis procedure (CONAN) which is designed to recognize land use patterns in high resolution remotely sensed data by analysis of the local frequency distribution of scene components (i.e. ground cover type classes). The procedure was tested with randomly generated synthetic data developed to simulate the frequency distribution of cover type components for four land use classes. It was found that the accuracy in discriminating between the four test classes depends upon the size of the pixel neighborhood used to compute the component frequency distribution.

Stephen W. Wharton | S. Wharton

[1] Geoffrey H. Ball,et al. ISODATA, A NOVEL METHOD OF DATA ANALYSIS AND PATTERN CLASSIFICATION , 1965 .

[2] Morris Goldberg,et al. A Clustering Scheme for Multispectral Images , 1978, IEEE Transactions on Systems, Man, and Cybernetics.

[3] A. Pentland,et al. Blob - An unsupervised clustering approach to spatial preprocessing of MSS imagery , 1977 .

[4] David A. Landgrebe,et al. The development of a spectral-spatial classifier for earth observational data , 1980, Pattern Recognit..

[5] P. M. Narendra,et al. A non-parametric clustering scheme for landsat , 1977, Pattern Recognit..

[6] Howard Jay Siegel,et al. Contextual Classification of Multispectral Remote Sensing Data Using a Multiprocessor System , 1980, IEEE Transactions on Geoscience and Remote Sensing.

[7] Azriel Rosenfeld,et al. A Relaxation Method for Multispectral Pixel Classification , 1980, IEEE Transactions on Pattern Analysis and Machine Intelligence.