Hyperspectral data dimensionality-reduction method based on sparse and low-rank representation graph

Abstract

Provided is a hyperspectral data dimensionality-reduction method based on a sparse and low-rank representation graph, sparse representation characteristics and low-rank representation characteristics with structures for maintaining global data are obtained via the L1 norm, and the method maintains the low-rank characteristics of a graph via a nuclear norm. The method includes following technical contents: 1) certain data is selected from original hyperspectral data as a training sample; 2) the sparse and low-rank representation graph of the selected training sample is constructed; and 3) an optimal projection matrix is sought via an optimization criterion so that the characteristics of the graph constructed in step 2 are maintained in a low-dimensional manifold space after projection. The characteristics of sparse and low-rank representation between sampling points are learned in an original space, data is projected to the low-dimensional manifold space via a conversion projection matrix, and the characteristics of sparse and low-rank representation between the sampling points can also be maintained.

Description

technical field [0001] The invention relates to a hyperspectral data dimensionality reduction method based on a sparse and low-rank representation graph, which belongs to the technical field of data processing and is suitable for dimensionality reduction and classification of hyperspectral data and reduction of band redundancy. Background technique [0002] In the field of hyperspectral image processing, due to the high-dimensional characteristics of hyperspectral data and the high correlation between each band, data dimensionality reduction plays an important role. The purpose of data dimensionality reduction is to reduce the computational complexity and improve the classification accuracy by reducing the feature dimension. Band selection and dimensionality reduction projection techniques are two main strategies for data dimensionality reduction. Band selection is a technology that directly extracts a few features from the original features according to some optimal criter...

AI Technical Summary

Problems solved by technology

However, the disadvantage of sparse representation is that only sparse sample points can be found, and there is no global constraint, so the global characteristics of the original data are lost in the low-dimensional manifold space.

Images