Hyperspectral Image Classification Method Combining Spectral, Spatial and Hierarchical Information

Abstract

The invention discloses a hyperspectral remote sensing image classification technology that combines spectrum, space and hierarchical structure information, including inputting hyperspectral images to be classified, extracting reference data sample sets; selecting training sample sets for supervised classification; and adopting principal component analysis for original images Extract the first three principal component bands of the image, use the extended morphology method to extract the spatial feature vector; use Markov feature selection to reduce the dimensionality of the original image, use the algebraic multi-grid method to construct the image pyramid, and use the hierarchical segmentation method to obtain multi-layer segmentation Results; the original spectral eigenvectors, spatial eigenvectors and hierarchical structure eigenvectors are combined to construct a multi-kernel matrix; the support vector machine method is used to calculate the difference of the kernel matrix to obtain the final category attribute labels; the final image classification map is output. The invention provides an effective classification method, which can fully extract and mine image information, and can effectively improve the classification accuracy of hyperspectral images.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a hyperspectral image classification method combining spectral, spatial and hierarchical information. Background technique [0002] Compared with multispectral remote sensing images, hyperspectral remote sensing images have richer information, which can accurately reflect the attribute differences between different types of ground objects, realize accurate extraction and identification of ground objects, and provide more accurate hyperspectral remote sensing image analysis. Lay a good foundation with industry applications. However, hyperspectral images have high dimensionality, high band correlation, noise and mixed pixels, and limited ground truth data references, which bring great challenges to the analysis and processing of hyperspectral remote sensing information. The single-kernel image classification method based on spectral features only determines the classifica...

Images