Remote sensing image classification method based on space spectrum capsule generative adversarial network

Abstract

The invention discloses a remote sensing image classification method based on a space spectrum capsule generative adversarial network model. The method comprises the following main steps: 1, creatinga generative adversarial network model; 2, determining a sample set; 3, training a generative adversarial network model by adopting the sample set in the step 2; 4, verifying the accuracy of the model; and 5, inputting a to-be-classified hyperspectral remote sensing image into the trained generative adversarial network model to obtain a classification result. According to the method, spectral information and spatial information are fully utilized, detail features such as relative positions of samples can be accurately modeled, and the classification precision and the classification efficiencyare greatly improved.

Description

technical field [0001] The invention belongs to image information processing technology, in particular to a remote sensing image classification method based on spatial spectrum capsule generation confrontation network. Background technique [0002] A hyperspectral image is a high-dimensional image collected by a spectral imager. It contains hundreds of spectral channels, so each pixel is a continuous and high-dimensional spectral curve. Specific bands can be selected or extracted to highlight target features as required. The hyperspectral imager detects the two-dimensional geometric space information and one-dimensional spectral information of the target at the same time, so the hyperspectral data has the structure of "image cube", reflecting the characteristics and advantages of "integration of map and spectrum". At present, hyperspectral images have been widely used in agriculture, military affairs, astronomy and other fields. [0003] The most prominent feature of hypers...

AI Technical Summary

Problems solved by technology

[0007] In order to solve the problem that the convolutional neural network does not comprehensively utilize spectral information and spatial information, and thus cannot accurately model detailed features such as the relative position of samples, resulting in poor classification accuracy and low classification efficiency due to large capsule network parameters, the present invention proposes A Remote Sensing Image Classification Method Based on Spatial Spectral Capsule Generative Adversarial Network Model

Images