Hyperspectral image classification method combining 3D/2D convolutional network and adaptive spectral unmixing

Abstract

The invention relates to a hyperspectral image classification method combining a 3D/2D convolution network and adaptive spectral unmixing, and the method comprises the steps: building a network modelthrough employing a 3D/2D dense connection network and a plurality of intermediate classifiers, and enabling the adaptive spectral unmixing to serve as the supplement of a network classification result. The design of multiple intermediate classifiers with early exit mechanisms enables the model to facilitate classification by using adaptive spectral unmixing, which brings considerable benefits tocomputational complexity and final classification performance. Besides, the invention further provides a 3D/2D convolution based on the spatial spectrum characteristics, so that the three-dimensionalconvolution can contain less three-dimensional convolution, and meanwhile, more spectral information is obtained by utilizing the two-dimensional convolution to enhance characteristic learning, so that the training complexity is reduced. Compared with an existing hyperspectral image classification method based on deep learning, the hyperspectral image classification method is higher in calculationefficiency and higher in precision.

Description

technical field [0001] The invention relates to a hyperspectral image classification method aimed at low spatial resolution. The method is a hyperspectral image classification combining 3D / 2D convolution network and adaptive spectral unmixing, and belongs to the field of image processing. Background technique [0002] Hyperspectral remote sensing images have high spectral resolution, multiple imaging bands, and large amounts of information, and are widely used in remote sensing applications. Hyperspectral image classification technology is a very important content in hyperspectral image processing technology. The rich spectral resolution in hyperspectral can improve the ability to accurately distinguish ground objects. While rich spectral resolution is useful for classification problems, the price of rich spectral resolution is lower spatial resolution. Due to the low spatial resolution, the spectral signature of each pixel will consist of a mixture of different spectra. ...

AI Technical Summary

Problems solved by technology

Therefore, the mixed pixels caused by low spatial resolution is one of the main obstacles affecting the accuracy of hyperspectral classification

Images