A hyperspectral image classification method based on deep learning space-spectral joint network

Abstract

The invention discloses a hyperspectral image classification method based on a deep learning space-spectrum joint network. First, the original hyperspectral image is divided into data, and then a small amount of label data is used to train the deep learning space-spectrum joint network. The input hyperspectral original The image will be extracted by the bidirectional long and short-term memory model with attention mechanism and the 1D hole convolutional neural network at the same time to obtain the final spectral feature map; the input image will be processed by data normalization, and then PCA reduction will be performed. dimension, extract the input features, and send the input features into the multi-scale and multi-layer filter convolutional network to extract the spatial features, and then process them through the global average pooling layer to obtain the final spatial feature map; finally, combine the trained network parameters. sort. In the present invention, the spectral dimension feature and the spatial feature are processed separately, so that a richer and more effective spectral feature map and richer feature expression can be obtained, and the classification accuracy is further improved.

Description

technical field [0001] The invention belongs to the field of remote sensing image processing, in particular to a hyperspectral image classification method based on a deep learning space-spectrum joint network. Background technique [0002] Hyperspectral remote sensing images can extract ground object information from hundreds of continuous spectral bands, which makes it have a strong ability to distinguish ground objects. In the past few decades, hyperspectral images have played an important role in military target detection, ocean monitoring, disaster prevention and so on. However, the identification and classification of hyperspectral images has always been a key issue in hyperspectral image analysis, and it plays a very important role in the advancement and development of hyperspectral remote sensing technology. The application potential of spectral remote sensing technology is of great significance. [0003] So far, researchers have proposed many methods to classify th...

AI Technical Summary

Problems solved by technology

However, the former has made band selection, which leads to a large loss of features, and the continuous context information in the spectral dimension will also be destroyed, while the latter cannot completely extract the entire spectral information, and both of them only use convolutional neural network extraction. feature, the network has outstanding advantages in extracting local related features, but cannot extract complete feature information

Images