Hyperspectral image classification based on set empirical mode decomposition of image features

Abstract

The invention relates to a hyperspectral image classification method based on set empirical mode decomposition of image features. The method comprises the following steps: S1, the original hyperspectral image is dimensionally reduced by using principal component analysis algorithm; S2, adaptive total variational filtering is performed on the obtained 20-dimensional principal components to reduce the sensitivity of noise and obtain rough contour features; 3, each spectral band is decomposed into serial components by using the set empirical mode decomposition, and further the features of the hyperspectral image are fused in the conversion domain; S4, the final classification of the processed image is performed by using a support vector machine classifier. The method can effectively enhance the contour feature of an image, and has better classification performance.

Description

technical field [0001] The invention relates to a hyperspectral image classification method, in particular to a hyperspectral image classification method based on the set empirical mode decomposition of image features. Background technique [0002] Hyperspectral imaging has greatly improved human cognition of land cover with high spectral resolution and wide spectral coverage of hyperspectral images. Hyperspectral images contain hundreds of spectral bands, which help to discover detailed spectral features hidden within narrow spectra. Hyperspectral image feature extraction is one of the hotspots in the field of hyperspectral image processing: such as principal component analysis, nuclear local Fisher discriminant analysis, chaos theory analysis. However, for the problem of difficult classification under image noise and spectral mixture, it is difficult to obtain satisfactory classification accuracy only by relying on spectral information. In recent years, the classificatio...

AI Technical Summary

Problems solved by technology

[0003] In view of the complex spectrum and spatial structure of hyperspectral images, the phenomenon of "same object with different spectra, same spectrum with different objects" is prone to occur, which makes it difficult to classify the same object and feature extraction. In order to solve the above problems, the present invention The following technical solutions are provided, including the following steps:

Images