Segmentation of multispectral high-resolution satellite imagery based on integrated feature distributions

Abstract

Texture features are useful for segmentation of high-resolution satellite imagery. This paper presents an efficient feature extraction method that considers the spatial and cross-band relationships of pixels in multispectral or colour images. The texture feature of an image region is represented by the joint distribution of two texture measures calculated from the first two principal components (PCs). Similarly, the spectral feature of the region is the joint distribution of greyscale pixel values of the two PCs. The texture distributions computed by a rotation invariant form of local binary patterns (LBP) and spectral distributions are adaptively combined into coarse-to-fine segmentation based on integrated multiple features (SIMF). The feasibility and effectiveness of the SIMF segmentation approach is evaluated with multispectral high-resolution satellite imagery and colour textured mosaic images under different conditions.