In this study the texture analysis is used to classify a high resolution QuickBird panchromatic image, in an urban area. The gray level co-occurrence matrix (GLCM) is computed to extract texture images. Texture measures are considered to be divided into three groups: Contrast, Orderliness and Statistic. The window size for the GLCM is determinated by semivariogram method. The principal component analysis (PCA) is then executed on texture images. The first few principal component bands (PC i), representing most of variations, are selected and assembled to panchromatic images. The selected PC bands, in conjunction with the original panchromatic image, are classified using an object-based approach to categorize pixels into three classes: buildings, roads and vegetation.
Texture Analysis for Urban Areas Classification in High Resolution Satellite Imagery
Pasquale Merola;Alessia Allegrini
2012
Abstract
In this study the texture analysis is used to classify a high resolution QuickBird panchromatic image, in an urban area. The gray level co-occurrence matrix (GLCM) is computed to extract texture images. Texture measures are considered to be divided into three groups: Contrast, Orderliness and Statistic. The window size for the GLCM is determinated by semivariogram method. The principal component analysis (PCA) is then executed on texture images. The first few principal component bands (PC i), representing most of variations, are selected and assembled to panchromatic images. The selected PC bands, in conjunction with the original panchromatic image, are classified using an object-based approach to categorize pixels into three classes: buildings, roads and vegetation.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.