Hyperspectral image space spectrum classification method and device considering spectral importance

Abstract

The invention discloses a hyperspectral image space spectrum classification method and device considering spectral importance, and the method comprises the steps: calculating the spectral feature importance of a given hyperspectral remote sensing image with hundreds of spectral bands through a random forest; customizing a spectral weight characteristic kernel function, modeling the relative effectof each spectral band in classification by utilizing the extracted spectral characteristic weight, giving a larger weight to the band which is more beneficial to ground object category identification, and improving the spectral discrimination capability of the ground object; constructing spectral unitary potential energy and spatial binary potential energy under a unified framework of a conditional random field, considering spectral characteristic weights in the spectral unitary potential energy, integrating a spectral weight kernel function to improve the distinguishing capability of different spectral characteristics in a nonlinear space, and modeling the spatial correlation of ground objects through the spatial binary potential energy. The beneficial effects of the method are that themethod reduces the impact on classification from an unimportant waveband, improves the discrimination of types, and improves the classification effect.

Description

technical field [0001] The invention belongs to the field of hyperspectral remote sensing image processing, and in particular relates to a hyperspectral image spatial spectrum classification method and device taking into account the importance of the spectrum. Background technique [0002] Hyperspectral remote sensing images have the unique advantages of high spectral resolution and map-spectrum integration, which can provide diagnostic spectral features of different ground objects, and are an important data source for classification and recognition of ground objects. In recent years, with the rapid development of hyperspectral technology, hyperspectral remote sensing images with high spatial resolution (double high remote sensing images) have begun to emerge, such as the domestic Tiangong-1 satellite, aviation ROSIS, CASI, and drones. The spatial resolution of hyperspectral imagery has reached meter level, or even submeter level. This kind of hyperspectral image with high ...

AI Technical Summary

Problems solved by technology

However, the current spatial spectral classification method based on spatial context information, represented by the random field model, focuses on the modeling of spatial information, and lacks a learning mechanism for sensitive spectral information of ground objects, so it is difficult to use spatial information to suppress classification noise while maintaining local classification. Details

Images