Sparse representation and empty spectrum Laplace figure based hyperspectral data dimension reduction method

Abstract

The invention discloses a sparse representation and empty spectrum Laplace figure based hyperspectral data dimension reduction method which mainly aims at solving the problems that the traditional manifold learning information is single and large-scale data are difficult to be processed. The sparse representation and empty spectrum Laplace figure based hyperspectral data dimension reduction method comprises the steps of step 1, selecting a certain amount of data from large-scale hyperspectral data to serve as a training sample; step 2, performing construction of an empty spectrum Laplace figure on the training sample; step 3, performing characteristic decomposition on a Laplacian matrix to obtain the low-dimension representation of the training sample; step 4, constructing a high-dimensional dictionary and a low-dimensional dictionary through the training sample and the low-dimension representation of the training sample; step 5, calculating sparse representation coefficients of remaining hyperspectral data on the high-dimensional dictionary; step 6, performing multiplication on the sparse representation coefficients and the low-dimensional dictionary to obtain the low-dimension representation of the remaining data; step 7, integrating the low-dimension representation of the training sample and the remaining data to obtain complete dimension reduction data. According to the sparse representation and empty spectrum Laplace figure based hyperspectral data dimension reduction method, the effect of the manifold dimension reduction is improved and the large-scale hyperspectral data can be processed.

Description

technical field [0001] The invention belongs to the technical field of data processing, and relates to pre-processing of hyperspectral data. The main purpose is to reduce the dimensionality of hyperspectral data, thereby reducing the computational complexity of later data processing methods, and at the same time improving its performance as much as possible. This method can be applied to large-scale hyperspectral data clustering or classification. Background technique [0002] Data dimensionality reduction processing plays a big role in data processing. Many data with too high dimensionality will be dimensionally reduced before processing. More useful features to improve the processing effect of the post-algorithm. Spectral data With the continuous improvement of the spectral resolution of imaging equipment, the dimensionality of data is also getting higher and higher, and data dimensionality reduction is essential. At the same time, with the development of equipment, the s...

AI Technical Summary

Problems solved by technology

This method handles the classification problem very well, but the parameters in the heat kernel used for weight calculation have a great influence on the embedding structure

[0008] The above methods have two unified defects: (1) The very important step in these methods is the construction of the graph. When the data scale is very large, the storage of the graph and the later calculation are very difficult. The general manifold learning The method cannot handle large-scale data; (2) The common manifold learning method does not take into account the spatial structure existing in the hyperspectral data, but simply considers the neighborhood relationship between the spectra, resulting in dimensionality reduction of the hyperspectral data. Unsatisfactory effect

Images