Hyperspectral data dimensionality reduction method based on ultra-pixel and maximum boundary distribution

Abstract

The invention discloses a hyperspectral data dimensionality reduction method based on ultra-pixel and maximum boundary distribution, which overcomes a defect of high inter-band correlation for lack of enough spatial and spectral domain information in the prior art. The hyperspectral data dimensionality reduction method comprises the steps of (1) dividing a sample set; (2) generating a regular matrix; (3) generating a judgment matrix; (4) solving an optimal projection matrix; and (5) carrying out projection dimensionality reduction. The hyperspectral data dimensionality reduction method has the advantages of abilities of maintaining the consistency in space and spectrum of neighboring samples and reducing redundant bands, and can be applied to dimensionality reduction of hyperspectral remote sensing images.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to a hyperspectral data dimensionality reduction method based on superpixels and maximum boundary distribution in the technical field of information extraction and machine learning. The invention can be used to reduce dimension and classify hyperspectral data, reduce redundant bands, and more accurately determine different types of ground objects in hyperspectral data. Background technique [0002] At present, in the field of hyperspectral remote sensing images, methods for dimensionality reduction of hyperspectral data are usually divided into two categories: feature extraction and feature selection methods. The feature extraction method uses the original data to extract its characteristic parameters, compresses the band through mathematical transformation, and projects the data to a low-dimensional space. Common methods include principal component analysis and linea...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that because this method only uses the spectral domain information between hyperspectral data, it lacks the spatial domain information of hyperspectral data, which affects the classification and recognition rate of hyperspectral images.

The disadvantage of this method is that due to the lack of spectral domain information of adjacent samples in hyperspectral data, band correlation and data information cannot be combined, so under the condition of maximum information, the correlation between bands will be large. Affect classification recognition rate

Images