Hyperspectral remote sensing image classification method and system based on 3D gabor feature selection

Abstract

The present invention is suitable for hyperspectral remote sensing image classification, and provides a hyperspectral remote sensing image classification method based on three-dimensional Gabor feature selection. The steps include: A, generating a three-dimensional Gabor filter according to the set frequency and direction parameter values; The remote sensing image is convolved with the three-dimensional Gabor filter to obtain the three-dimensional Gabor feature; C, select a number of three-dimensional Gabor features that meet the requirements for various classification contributions from the three-dimensional Gabor feature; D, use the selected three-dimensional Gabor feature to pass Multi-task sparse classification method to classify hyperspectral remote sensing images. The present invention is based on the three-dimensional Gabor feature, and the three-dimensional Gabor feature used contains signal-rich local change information, and the feature expression ability is strong; the three-dimensional Gabor feature is selected through the Fisher discriminant criterion, which makes full use of the hidden high-level semantics between the features, and removes redundant information. The time complexity of classification is reduced; further, using sparse coding, combining 3D Gabor features and multi-tasks, the classification accuracy is greatly improved.

Description

technical field [0001] The invention belongs to the field of data classification, in particular to a hyperspectral remote sensing image classification method and system based on three-dimensional Gabor feature selection and multi-task sparse representation. Background technique [0002] Hyperspectral remote sensing images are multispectral image data acquired by remote sensing sensors from objects of interest on the ground in the visible, near-infrared, mid-infrared and thermal infrared bands of the electromagnetic spectrum. Hyperspectral remote sensing images contain rich triple information of space and radiation spectrum, and show good results in the classification of ground materials. Traditional classification methods use commonly used classifiers (K nearest neighbors, support vector machines) to classify directly on hyperspectral remote sensing images, which cannot meet the actual classification effect. In view of the spatial and spectral three-dimensional structure of...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is to provide a hyperspectral remote sensing image classification method and system based on multi-task sparse representation based on three-dimensional Gabor feature selection, aiming at solving a large number of problems in the classification of hyperspectral remote sensing images in the prior art. Redundant information that is not conducive to classification reduces classification accuracy and increases classification time complexity

Images