Hyperspectral reflectance data spectrum characteristics extracting method based on global sensitivity analysis

Abstract

The invention relates to a hyperspectral reflectance data spectrum characteristics extracting method based on global sensitivity analysis. The method comprises the following steps of (1) sensitivity analysis: calculating sensitivity of the hyperspectral reflectance data on each waveband position by utilizing a global sensitivity analysis method; (2) characteristic waveband selection: selecting a characteristic waveband according to the sensitivity analysis result; (3) canonical transformation: calculating a characteristic vector by utilizing a canonical transformation method when the divisibility of the class is maximal; (4) spectrum characteristic establishment: establishing a first canonical axis to be used as a spectrum characteristic by utilizing the linear combination of the characteristic waveband reflectance value and the elemental value corresponding to the characteristic vector. Compared with the traditional hyperspectral reflectance data spectrum characteristics extracting method, the spectrum characteristic established by adopting the method enables the divisibility of different classes to be maximized while the data dimension is reduced, and the method is particularly applicable to the remote sensing application problem such as crop stress detection, target recognition, ground feature classification and the like, and has wide prospect in the technical field of the hyperspectral reflectance data processing and application.

Description

technical field [0001] The invention relates to a method for extracting spectral features of hyperspectral reflectance data, and belongs to the technical field of hyperspectral reflectance data processing methods and applications. Background technique [0002] With the emergence of spaceborne hyperspectral resolution sensors and the development of quantitative remote sensing models, the quantitative application of hyperspectral remote sensing data has attracted more and more attention. Compared with traditional low-resolution remote sensing data, hyperspectral remote sensing data has more bands and provides a larger amount of data, so it has broader application potential. However, the improvement of spectral resolution and the increase of data volume also bring challenges to the quantitative application of remote sensing data. Generally speaking, although the increase in the amount of data helps to interpret more useful information, it also brings a lot of redundant informa...

AI Technical Summary

Problems solved by technology

One is that the band selection process is based on training samples, resulting in the selected bands depending on the characteristics of a specific training data set, so the universality and scalability are poor; some methods do not include the band selection process, which makes the feature extraction a large amount of calculation and consumes a lot of time. Time

The second is that for specific problems, such as crop stress detection, target recognition and ground object classification, the separability between different categories is not considered in the process of constructing spectral features, so the effect in dealing with these problems is poor

Images