Remote sensing image landslide automatic detection method based on three-dimensional space-channel attention mechanism

A technology of three-dimensional space and remote sensing images, applied in neural learning methods, image data processing, computer components, etc., to achieve good recognition results, strong robustness, and strong reusability

Active Publication Date: 2020-06-02
WUHAN UNIV
View PDF5 Cites 18 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But these studies are still in their infancy, and only

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
  • Remote sensing image landslide automatic detection method based on three-dimensional space-channel attention mechanism
  • Remote sensing image landslide automatic detection method based on three-dimensional space-channel attention mechanism
  • Remote sensing image landslide automatic detection method based on three-dimensional space-channel attention mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0033] First construct the convolutional neural network (step 1) based on the three-dimensional space-channel attention mechanism according to the method of the present invention, as follows Figure 4 The attention module shown is embedded into an existing residual network. During the training process of the network, the input of the attention module is the intermediate feature map inside the network, and the output is the weighted feature map with the same size as the input, and then the weighted feature map is input into the original network to continue the calculation . In simple terms, this process is to adaptively weight an intermediate calculation result inside the network. Figure 4 The cubes in represent different feature maps, Represents a point-by-point multiplication operation, Indicates a point-by-point addition operation, indicates an activation operation. After completing the construction of the network, it is necessary to obtain training sample data 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 remote sensing image landslide automatic detection method based on a three-dimensional space-channel attention mechanism. The method comprises the following steps: manuallylabeling remote sensing images, cutting landslide samples to construct a sample library, training a convolutional neural network based on a three-dimensional space-channel attention mechanism, and learning features of various landslide instances on the remote sensing images; and carrying out landslide detection on the new remote sensing image by using the trained network model to realize automaticand accurate recognition of a landslide target on the remote sensing image. The method has the advantages that the robustness is high, and compared with an original backbone network and other existing attention methods, the method is more suitable for a landslide detection task on a remote sensing image; for the condition containing various interference factors, the method has better interferenceresistance, and higher recognition accuracy can be obtained; the method can be used for preventing landslide disasters, reconstructing after disasters, updating landslide databases and the like.

Description

technical field [0001] The invention relates to an automatic detection method of landslides in remote sensing images based on a three-dimensional space-channel attention mechanism. The method can effectively automatically discover and identify landslides from remote sensing images, and can be used for landslide disaster prevention, post-disaster reconstruction, and landslide database update. . Background technique [0002] Landslide is a common natural disaster, which poses a serious threat to the natural environment and the safety of people's lives and property. Landslide detection and identification can provide important data support for the prevention and control of landslide disasters. In the early days, landslide data were mainly obtained through on-the-spot surveys by field workers. This landslide detection method is time-consuming, laborious and dangerous. With the in-depth study of landslide properties, some automatic landslide detection methods have been gradually ...

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): G06K9/00G06K9/62G06N3/04G06N3/08G06T17/05
CPCG06N3/08G06T17/05G06V20/13G06N3/048G06N3/045G06F18/241
Inventor 季顺平余大文
Owner WUHAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap