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

Automatic construction method of optical satellite image feature matching deep learning training sample set

A training sample set and feature matching technology, applied in the field of remote sensing information processing, can solve the problems of poor versatility, high cost, low efficiency of training sample set construction, etc., and achieve the effect of improving efficiency, reducing cost and professional threshold, and high practical value.

Active Publication Date: 2020-11-17
PLA PEOPLES LIBERATION ARMY OF CHINA STRATEGIC SUPPORT FORCE AEROSPACE ENG UNIV
View PDF2 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the technical problems of low training sample set construction efficiency, high cost, and poor versatility when using deep learning methods for optical satellite image feature matching

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
  • Automatic construction method of optical satellite image feature matching deep learning training sample set
  • Automatic construction method of optical satellite image feature matching deep learning training sample set
  • Automatic construction method of optical satellite image feature matching deep learning training sample set

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0052] In order to verify the effectiveness of the present invention, a simulated satellite image was generated using a reference image and a digital elevation model in a certain area in Henan, and a sample set of feature matching points with the same name was constructed. The specific implementation process is as follows:

[0053] 1. If image 3 , 4 As shown in , collect reference images and digital elevation models of a certain area in Henan;

[0054] 2. If Figure 5 As shown, the reference images and digital elevation model are used to simulate the generation of optical linear array pushbroom satellite images in this area;

[0055] 3. With the assistance of the internal and external orientation elements of the simulated image, a grayscale-based image matching method is used to match a number of points with the same name on the reference image and the simulated image, that is, step 2-step 4 of the present invention, such as Figure 6-9 shown

[0056] 4. Using the above m...

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 belongs to the technical field of remote sensing information processing, and provides an automatic construction method of an optical satellite image feature matching deep learning training sample set, which performs full-link simulation on a satellite imaging process, and obtains a simulation image and corresponding internal and external orientation elements by using a reference image and a digital elevation model; feature extraction is made on the reference image and the simulation image by using a gray scale-based or feature-based method to obtain a plurality of feature points;a reasonable search range is set and whether the search range is a homonymous point or not is veridied; feature points on the reference image are calculated one by one, and construction of a positivesample set is completed; a certain number of feature points on the reference image and the simulation image are randomly selected to construct a negative sample set; the optical satellite image feature matching deep learning training sample set automatic construction method without manual annotation is realized so that the efficiency and the reliability of sample set construction can be greatly enhanced and the cost and the professional threshold can be reduced.

Description

technical field [0001] The invention belongs to the technical field of remote sensing information processing, and in particular relates to a method for automatically constructing an optical satellite image feature matching deep learning training sample set. Background technique [0002] The feature matching of optical remote sensing satellite images is an important prerequisite for image registration and block adjustment connection point extraction. Because different satellite images have large differences in imaging methods, ground resolution, time phase, and lighting conditions, it is difficult to match satellite image features. Traditional grayscale-based or feature-based matching methods have a high accuracy rate in practice Low, low success rate and poor reliability. At present, the image feature matching method based on deep learning shows great development potential, and solves the problems existing in the traditional method to a large extent (document: Fan Dazhao, D...

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): G06K9/00G06K9/46G06K9/62
CPCG06V20/13G06V10/44G06V10/751
Inventor 薛武王鹏夏鲁瑞钟灵毓倪蕾张旭李森
Owner PLA PEOPLES LIBERATION ARMY OF CHINA STRATEGIC SUPPORT FORCE AEROSPACE ENG UNIV
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