Fully-constrained least square linear spectrum hybrid analysis method of hyperspectral image

Abstract

The invention provides a fully-constrained least square linear spectrum hybrid analysis method of a hyperspectral image. The method comprises the following steps of: 1) initially analyzing a mixing ratio; 2) screening an end member; 3) secondarily analyzing the mixing ratio; 4) inputting a mixed pixel to be analyzed, namely p; and 5) inputting d types of d end member arrays. The invention provides a new first come last served-linear spectral mixture analysis (FCLS-LSMA) analysis method in a single end member mode. The method has the advantages of high speed and optimal analysis effect theory.

Description

technical field The present invention relates to a spectral mixing analysis method of a hyperspectral image, in particular to a fully constrained least squares linear spectral mixing analysis method under a hyperspectral image single-end member mode (that is, each type has only one end member information), It belongs to the technical field of remote sensing information processing. Background technique The spatial resolution of hyperspectral images is generally low, which leads to the widespread existence of mixed pixels, that is, a pixel may be a mixture of several categories. For such pixels, it is inaccurate to attribute them to any category according to the general classification method. The technology of analyzing the proportion of each type of component in the mixed pixel is called spectral mixing ratio analysis, which is one of the most basic and important contents of hyperspectral data analysis. In essence, it is a more accurate classification technology. . The res...

AI Technical Summary

Problems solved by technology

Although the distance calculation only includes matrix multiplication, due to the high dimensionality of hyperspectral data, the amount of matrix multiplication required at this time is still relatively large

On the other hand, when the mixed pixel falls inside the convex polyhedron formed by each end member, the results obtained by the method proposed by Geng Xiurui and Luo Wenfei meet the full constraints; and when the mixed pixel falls outside the convex polyhedron, the above method The full constraints cannot be fully satisfied, and the ordinary full constraint adjustment method (that is, set the components less than 0 to 0, the components greater than 100% to 100%, and then divide each component by the sum of all components) does not meet The meaning of least squares

Images