. .  , 2003, . 24, . 9, 1949–1956 Technical note Classification of off-diagonal points in a co-occurrence matrix D. B. HANN, Department of Mechanical Engineering, King’s College London, University of London, Strand, London, WC2R 2LS, UK; e-mail: david.hann@kcl.ac.uk A. M. S. SMITH* Department of Geography, King’s College London, University of London, Strand, London, WC2R 2LS, UK; e-mail: alistair.smith.kcl@kcl.ac.uk and A. K. POWELL Department of Physics, King’s College London, University of London, Strand, London, WC2R 2LS, UK; e-mail: kp@maxwell.ph.kcl.ac.uk (Received 26 November 2001; in final form 15 March 2002 ) Abstract. The classification of the off-diagonal points within a typical grey level co-occurrence matrix (GLCM ) is discussed through the application of an intuitive nearest peak and a boundary rule method. Both approaches are applied to a synthetic image consisting of five regions with varying amounts of added random noise and also to an image containing three Brodatz textures of different standard deviation. The two approaches correctly identify the majority of the internal region pixels. However, the nearest peak method is shown to produce serious misclassifications at the region boundaries in the form of bands of additional regions. The boundary rule method does not show this characteristic. The overall classification accuracy and the k hat statistic were used to test the performance of each technique. 1. Introduction The determination of image features is a subject undergoing much research in the analysis of Earth observation imagery. One technique that has recently been receiving increasing attention in the remote sensing community is the use of co-occurrence matrices for the identification of textural classes within an image (Carlson and Ebel 1995, Karathanassi et al. 2000, Kiema 2002, Smith et al. in press). Classification techniques based on co-occurrence matrices use either the inherent textural properties of image entropy, homogeneity, dissimilarity, contrast, etc., (Haralick and Shanmugam 1974, Franklin et al. 2001, Kiema 2002) or work by * Corresponding author. International Journal of Remote Sensing ISSN 0143-1161 print/ISSN 1366-5901 online © 2003 Taylor & Francis Ltd http://www.tandf.co.uk/journals DOI: 10.1080/01431160210154074