Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Hyperspectral image classification method based on spectral space attention fusion and deformable convolutional residual network

A hyperspectral image and spectral space technology, applied in the field of hyperspectral image classification, can solve the problems of insufficient extraction of spectral and spatial features, low accuracy of hyperspectral image classification, etc., to enhance the ability of spatial spectral feature representation and contribute to The effect of accurate classification and enhanced generalization ability

Active Publication Date: 2021-09-07
QIQIHAR UNIVERSITY
View PDF8 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to solve the problem of insufficient spectral and spatial feature extraction in existing hyperspectral image classification because hyperspectral images contain rich information, and overfitting under small samples, resulting in low classification accuracy of hyperspectral images. problem, and propose a hyperspectral image classification method based on spectral-spatial attention fusion and deformable convolutional residual network

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Hyperspectral image classification method based on spectral space attention fusion and deformable convolutional residual network
  • Hyperspectral image classification method based on spectral space attention fusion and deformable convolutional residual network
  • Hyperspectral image classification method based on spectral space attention fusion and deformable convolutional residual network

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0034] Specific implementation mode 1: In this implementation mode, the specific process of the hyperspectral image classification method based on spectral spatial attention fusion and deformable convolutional residual network is as follows:

[0035] Spectral information and spatial information are equally important for hyperspectral image classification. Many studies have shown that considering both feature extraction methods is much better than relying on only one of them. According to different stages of spectral feature fusion, these methods can be classified into three categories: preprocessing-based networks, ensemble networks, and post-processing-based networks.

[0036] Preprocessing based network

[0037] The classification process based on preprocessing usually includes two stages: 1) spectral-spatial feature extraction and fusion stage; 2) the extracted features are classified by different classifiers (such as SVM). The first stage is critical to determine the per...

specific Embodiment approach 2

[0054] Specific embodiment 2: The difference between this embodiment and specific embodiment 1 is that in the step 2, SSAF-DCR based on spectral spatial attention fusion and deformable convolution residual network includes:

[0055] The first input layer, the first unit, the second output layer, the sixth batch of normalization layer, the first hidden layer Dropout, the third input layer, the second unit, the thirteenth three-dimensional convolution, the thirteenth batch of normalization layer, second hidden layer Dropout, fifth input layer, third unit, global average pooling layer, fully connected layer;

[0056] The first unit includes: a first three-dimensional convolutional layer, a spectrally dense block, a fifth normalization layer, a fifth PReLU activation layer, a sixth three-dimensional convolutional layer, and a spectral attention block;

[0057] The spectrally dense block includes: the first batch of normalization layers, the first PReLU activation layer, the second...

specific Embodiment approach 3

[0064] Specific embodiment three: the difference between this embodiment and specific embodiment one or two is that the SSAF-DCR connection relationship based on spectral spatial attention fusion and deformable convolutional residual network is:

[0065] The output layer of the first input layer is connected to the input of the first three-dimensional convolutional layer in the first unit, and the output of the first three-dimensional convolutional layer is respectively used as the input of the first batch of normalization layers in the spectral dense block, and the second batch of The input of the normalization layer, the input of the third batch of normalization layer and the input of the fifth batch of normalization layer;

[0066] The output of the first batch of normalization layers is connected to the input of the first PReLU activation layer, the output of the first PReLU activation layer is connected to the input of the second three-dimensional convolutional layer, and ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a hyperspectral image classification method, and concretely relates to a hyperspectral image classification method based on spectral space attention fusion and a deformable convolutional residual network. The objective of the invention is to solve the problem of low classification accuracy of hyperspectral images caused by insufficient spectrum and spatial feature extraction and overfitting under small samples due to the fact that the hyperspectral images contain abundant information in the existing hyperspectral image classification. The method comprises the following steps: 1, acquiring a hyperspectral image data set and a corresponding label vector data set; 2, establishing an SSAF-DCR network based on spectrum space attention fusion and a deformable convolution residual error; 3, inputting the x1, the x2, the Y1 and the Y2 into a network SSAF-DCR, and performing iterative optimization by adopting an Adam algorithm to obtain an optimal network; and 4, inputting x3 into the optimal network to carry out classification result prediction. The method is applied to the field of hyperspectral image classification.

Description

technical field [0001] The invention relates to a hyperspectral image classification method. Background technique [0002] Hyperspectral Images (HSIs) are images acquired by some spacecraft equipped with hyperspectral imagers. Each pixel of the image contains reflection information of hundreds of different bands, which makes this type of image suitable for many practical applications, such as military target detection, mineral exploration, agricultural production [1-4] Wait. In the field of hyperspectral image analysis and processing, including hyperspectral image classification, many research results have been obtained. However, there are still three problems in hyperspectral image classification: 1) In theory, the deeper the network layer, the higher the classification accuracy. However, for hyperspectral data with a huge amount of data, the network classification effect that is too deep and too complex is not ideal; 2) Compared with natural images, hyperspectral data s...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06K9/00G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06N3/048G06N3/045G06F18/23G06F18/24G06F18/214Y02T10/40
Inventor 石翠萍张甜雨王天毅
Owner QIQIHAR UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com