Hyperspectral classification method based on dual branch network

Abstract

The hyperspectral classification method based on the dual-branch network provided by the present invention, firstly, through the method of data resampling, it not only ensures that the number of samples of each category of input data is not constant and equal at each iteration in the training process, but also ensures that the number of samples in each category in the statistical Each type of sample participating in the training is balanced. This not only effectively alleviates the problem of sample imbalance in network learning, but also maintains the diversity of data; in order to extract multi-scale features of data, the present invention uses a double-branch network structure and performs semi-supervised learning through three training strategies. This not only expands the training set, but also greatly improves the classification accuracy compared with other classification methods through the integrated learning strategy. The hyperspectral classification method based on the dual-branch network proposed by the present invention not only outperforms other methods in performance, but also outperforms other methods in training efficiency.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a hyperspectral classification method based on a double-branch network. Background technique [0002] High-resolution hyperspectral images can not only display the geometric structure and spatial information of ground objects, but also contain rich spectral information. Therefore, hyperspectral images provide the basis for a wide range of applications, such as ground object recognition and classification, mineral exploration, precision agriculture, etc. For these applications, the most fundamental task is the classification of hyperspectral images. However, in the classification of hyperspectral images, there are still many challenges, such as, inter-class similarity, limited training data, sample imbalance, etc. [0003] To solve these problems, many researchers have proposed different solutions. Initially, Bigdeli et al. applied support vector machines (...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a hyperspectral classification method based on a dual-branch network, which solves the problem of poor classification ability of existing methods for classifying hyperspectral images

Images