Hyperspectral image classification method based on semi-supervised dictionary learning

Abstract

The invention discloses a hyperspectral image classification method based on semi-supervised dictionary learning, mainly solving the problem of high dimension of hyperspectral image and low classification precision in small sample status. The hyperspectral image classification method includes: expressing the pixel dot of the hyperspectral image via the spectral feature vector; selecting mark sample set, no-mark sample set and the test sample set; forming class label matrix with mark sample; forming Laplacian matrix without mark sample; using the alternating optimization strategy and the gradient descent method for solving the semi-supervised dictionary learning model; using the learned dictionary for coding the mark sample, no-mark sample and test sample; using the learned sparse code as the characteristic for classifying the hyperspectral image. The hyperspectral image classification method based on semi-supervised dictionary learning adopts the semi-supervised thought to obtain higher classification accuracy compared with the supervised learning method and is applied to the field of precision agriculture, vegetation investigation and military reconnaissance.

Description

technical field [0001] The invention belongs to the technical field of image processing, relates to semi-supervised learning and sparse representation methods, and is used for hyperspectral image classification problems under relatively small sample scenarios. Background technique [0002] Hyperspectral remote sensing technology was born in the 1980s. It combines imaging technology and spectral technology, and can obtain the radiation characteristics of ground objects of interest in tens to hundreds of narrow continuous bands from the ultraviolet to the near infrared of electromagnetic waves. An important cutting-edge technology for Earth observation. Compared with traditional spectral imaging technology, hyperspectral remote sensing not only has a higher number of bands and spectral resolution, but also has almost continuous bands, which can generate a continuous spectral curve for each pixel difference. , and spectrum triple information, which has the characteristics of s...

AI Technical Summary

Problems solved by technology

[0006] Although SVM has certain robustness to high-dimensional data, studies have shown that high-dimensionality still has a great impact on the classification performance of SVM; the semi-supervised method based on graph can improve the classification performance compared with the supervised algorithm, but it It directly deals with high-dimensional spectral features and cannot effectively improve the classification accuracy.

Images