A method for identification of moraine-covered glaciers combining polarization decomposition and topographic features

Abstract

The invention relates to a method for identifying surface moraine-covered glaciers combining polarization decomposition and topographic features. S Calculate the coherence matrix T , then through polarization decomposition and geocoding, the polarization decomposition features, the local incident angle, and the angle δ between the radar beam and the slope aspect are obtained. The principal component analysis of multiple polarization decomposition features is combined with the local incidence angle to synthesize the image to be classified. , δ is divided according to the unit of 60 degrees to obtain the vector files of small (S), medium (M) and large (L) areas. Second, segment the optical image to obtain the ground type sample vector, and crop the image to be classified to calculate the sample value. Finally, the SVM classification algorithm is performed on the images to be classified to identify the moraine-covered glaciers. Compared with the traditional method, the polarization decomposition feature used in the present invention is closely related to the physical properties of the glacier surface, and the partition classification can effectively reduce the misclassification caused by the terrain, further refine the classification of the glacier area and improve the identification accuracy.

Description

technical field [0001] The invention relates to a method for identifying a moraine-covered glacier on a SAR image based on polarization decomposition characteristics and angle information between a radar beam and a slope aspect, and belongs to the technical field of remote sensing geoscience applications. Background technique [0002] As the world's climate continues to change and extreme weather occurs frequently, people are paying more and more attention to the impact of climate on the cryosphere on Earth, especially glaciers, which are important indicators of climate change. Using remote sensing to identify glaciers is of great significance for understanding glacier resources and analyzing global climate change. However, optical images are restricted by weather conditions, and the glacier area is covered by clouds most of the year, and optical images can only identify part of the clean glaciers, and thermal infrared images are also affected by clouds. The mechanism of gl...

AI Technical Summary

Problems solved by technology

However, optical images are restricted by weather conditions, and the glacier area is covered by clouds for most of the year, and optical images can only identify part of the clean glaciers, thermal infrared images will also be affected by clouds

The mechanism of glacier recognition by optical images is the reflection of solar radiation on the surface of the glacier, but the phenomenon of different objects with the same spectrum and the same object with different spectra can easily lead to misclassification of glacier regions, especially the spectral characteristics of moraine-covered glaciers are similar to those of rocks

Images