Target detection method and system based on space-spectrum combination local preserving projection algorithm

Abstract

The invention discloses a target detection method and system based on a spatial-spectral combination local preserving projection algorithm. According to the method, an end member extraction step is added to an LPP algorithm, the adjacency relation between pixels is indirectly measured by calculating the similarity between the pixels and image end members, the influence of spectral fluctuation on the feature similarity is reduced, and the construction precision of an adjacency weight matrix is improved; on the basis, by adding the spatial domain adjacency weight matrix of the pixel, the effective sample points in each direction of the pixel can fully play a role, so that the dimension reduction data more accurately reflect the essential structure of the original hyperspectral image, and the detection precision of a target detection result can be improved.

Description

technical field [0001] The invention relates to the technical field of remote sensing image data processing and application, in particular to a target detection method and system based on a local-preserving projection algorithm based on space-spectrum combination. Background technique [0002] Dimensionality reduction of hyperspectral data is the first step in the processing and application of remote sensing image data, and it is also an important means and basis for people to obtain remote sensing information. At present, hyperspectral image dimensionality reduction methods are mainly divided into two categories: linear dimensionality reduction and nonlinear dimensionality reduction. Typical representatives of linear dimensionality reduction methods include principal component analysis (Principal Component Analysis, PCA) method and independent component analysis algorithm. The low-dimensional features obtained by these methods can better preserve the global characteristics...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a target detection method and system based on a space-spectrum combination local preservation projection algorithm, to solve the problem that the existing hyperspectral image dimensionality reduction method mainly uses the spectral information of the hyperspectral image in the dimensionality reduction process, ignoring other Spatial distribution information, which leads to the problem of low target detection accuracy

Images