Hyperspectral image classification method and device combining EMP features and TNT module

Abstract

The invention belongs to the technical field of hyperspectral image classification, and particularly relates to a hyperspectral image classification method and device combining EMP features and a TNT module, and the method comprises the steps: firstly, extracting the EMP features of a whole hyperspectral image through an extended morphological profile, and sequentially dividing a generated EMP cube into a plurality of plaques; expanding each plaque and performing linear transformation to obtain a plaque embedding and a plurality of pixel embedding; and finally, adding plaque embedding and pixel embedding into corresponding position codes respectively, and inputting obtained vectors into a deep network model containing L TNT modules together for classification. Compared with a support vector machine and other CNN deep learning models, the invention can obtain higher classification precision.

Description

technical field [0001] The invention belongs to the technical field of hyperspectral image classification, and in particular relates to a hyperspectral image classification method and device combining EMP features and a TNT module. Background technique [0002] Hyperspectral image classification is one of the most important links in hyperspectral image processing and analysis, and its accurate classification results can provide powerful data support for subsequent tasks. At present, hyperspectral image classification has been widely used in many fields such as precision agriculture, urban planning, and resource exploration. Hyperspectral images contain rich spectral information, and each pixel has an approximately continuous spectral curve, which provides the possibility for accurate classification and recognition of ground objects. However, the high-dimensional complexity of hyperspectral images and the correlation between bands have an impact on classification and recogni...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the problems that the deep learning model based on convolutional neural network (CNN) in the prior art is not good at modeling long-range dependencies and obtaining global context information, the present invention proposes a combination of EMP features and TNT (Transformer-iN- Transformer) module hyperspectral image classification method and device, the method first extracts the EMP features of the hyperspectral image, and then directly inputs the obtained EMP cube into the constructed deep network model based on the TNT module for end-to-end classification, improving high Classification Accuracy of Spectral Imagery

Images