Hyperspectral characteristic variable selection method

Abstract

The invention discloses a hyperspectral characteristic variable selection method and provides a characteristic forward selection support vector machine (SVM-BFFS) which is an embedded type forward sequence selection method base on the SVM maximum interval principle. The hyperspectral characteristic variable selection method is popularized to support the SVM-BFFS of a variety of classification problems by utilization of a one-to-one strategy. According to the hyperspectral characteristic variable selection method, stability of characteristics is ranked by exploring the internal relation between characteristic variables and the stability, and therefore the characteristic variables of infrared hyperspectra can be selected quickly and effectively. By application of the hyperspectral characteristic variable selection method, dimensionality of a hyperspectral image can be reduced, and therefore complexity of follow-up calculation is reduced, and a high accuracy rate of a built model is maintained.

Description

technical field [0001] The invention belongs to the technical field of image processing, relates to dimensionality reduction of high-dimensional data, is used for classification of hyperspectral remote sensing images, and in particular relates to a method for selecting hyperspectral characteristic variables. Background technique [0002] Hyperspectral remote sensing technology is one of the most important developments in the field of remote sensing in the 1980s, and it is also a cutting-edge technology in this field. Hyperspectral remote sensing technology uses imaging spectral scanners to simultaneously generate images of hundreds of bands for the observed object with nanoscale spectral resolution. Being able to record the continuous spectral information of the measured object, and having the characteristics of "unity of graphs and spectra", has made human beings take another big step forward in the field of remote sensing. Near-infrared hyperspectral is a kind of hyperspe...

AI Technical Summary

Problems solved by technology

The disadvantage is that the calculation is large, the running time is long, and it is easy to produce tight coupling with the selected regression classifier.

Support Vector Machine-Feature Recursive Vector Elimination (SVM-RFE), and the MFFS-SVM proposed in this paper are all embedded methods; the core principle of the feature selection method originally proposed by SVM-RFE to solve binary biological genes is based on the support The weight vector sorting of the vector machine performs cyclic recursive blanking on irrelevant variables. Although it performs well in binary classification, it does not have a good solution for multi-class problems.

Due to the post-order feature selection strategy, this method requires a lot of running time in the selection of high-dimensional hyperspectral feature variables.

Images