Transform-based remote sensing VHR image change detection method

An image change detection and change detection technology, applied in the field of remote sensing VHR image change detection based on Transformer, can solve the problems of low change detection accuracy, and achieve the effect of improving accuracy and enhancing correlation

Pending Publication Date: 2022-03-11
XIAN UNIV OF TECH
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a Transformer-based remote sensing VHR image change detection method, which solves the problem that the change detection accuracy rate in the change detection task of the prior art is low and cannot meet the technical requirements

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
  • Transform-based remote sensing VHR image change detection method
  • Transform-based remote sensing VHR image change detection method
  • Transform-based remote sensing VHR image change detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0019] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0020] refer to figure 1 , the image change detection method of the present invention, based on the Transformer principle, is implemented according to the following steps:

[0021] Step 1, build a preprocessing module, the input of the preprocessing module is the input image I of the dual-temporal remote sensing VHR 1 , I 2 , the size is H×W×3; the output of this preprocessing module is the feature map I 1-1 , I 2-1 , with a size of

[0022] refer to figure 2 , the main function of the preprocessing module is to input the input image I of the dual-temporal remote sensing VHR 1 , I 2 Perform data preprocessing, the structure of the preprocessing module is as follows: the input image I of dual-temporal remote sensing VHR 1 , I 2 (Input image) as the input image→Patch Partition layer (expression Conv4×4×48)→Linear Embedding layer (ex...

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 discloses a remote sensing VHR image change detection method based on Transform, and the method comprises the steps: 1, constructing a preprocessing module, inputting dual-time-phase remote sensing VHR input images I1 and I2, and outputting the dual-time-phase remote sensing VHR input images I1-1 and I2-1; 2, a Transform module is constructed, input data are the feature maps I1-1 and I2-1 output in the step 1, and output is extracted feature maps O1 and O2; 3, a change detection module is constructed, input data are the feature maps O1 and O2 output in the step 2, and an output image subjected to binarization processing after change detection and a binary image with the size being H * W are output. According to the method, based on the combination of the latest Transform model and the change detection task, the most advanced performance is obtained in the tasks of image classification, target detection, semantic segmentation and the like, and the method is applied to the change detection task to improve the change detection accuracy.

Description

technical field [0001] The invention belongs to the technical field of computer image processing, and relates to a Transformer-based remote sensing VHR image change detection method. Background technique [0002] Remote sensing VHR (very high resolution) images mainly refer to remote sensing images with high spatial resolution, which are measured by the size of the ground corresponding to the unit pixel, and the unit is meter. Compared with traditional medium and low resolution remote sensing images, remote sensing VHR images can more clearly express the spatial correlation and corresponding feature distribution of ground objects, which can provide more detailed ground objects and targets for various remote sensing image tasks. The structure provides good conditions for the wider application of remote sensing images. Image change detection methods have been continuously developed in recent years, from simple mathematical operations to the fusion of multiple methods, from pi...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/00G06N3/04G06N3/08
CPCG06T7/0002G06N3/08G06T2207/10032G06T2207/20081G06T2207/20084G06N3/047G06N3/048G06N3/045G06N3/044
Inventor 孙帮勇刘秦森王宇通陈金岭匡璐璐杨宗辉
Owner XIAN UNIV OF TECH
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