Remote sensing image matching method and device

A remote sensing image and matching method technology, which is applied in neural learning methods, instruments, biological neural network models, etc., can solve the problem that heterogeneous remote sensing image matching methods cannot be applied to large scenes of environmental changes, etc., to improve data processing efficiency and improve matching Performance, Accuracy Improvement Effects

Pending Publication Date: 2020-07-31
PLA STRATEGIC SUPPORT FORCE INFORMATION ENG UNIV PLA SSF IEU
View PDF3 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to provide a remote sensing image matching method and device to solve the problem that the current heterogeneous remote sensing image matching method cannot be applied to large scenes of environmental changes

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 matching method and device
  • Remote sensing image matching method and device
  • Remote sensing image matching method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0052] method embodiment

[0053] The present invention firstly inputs the image to be matched into the trained improved VGG16 network, extracts the conv4_3 convolutional layer in the improved VGG16 network to generate 512 channel feature maps, uses two strategies of channel maximum and local maximum to select key points, and Extract the 512-dimensional descriptor of the key point position on the feature map to complete the feature extraction and description; then use the fast nearest neighbor search (FLANN) feature matching, in order to solve the problem of many mismatching points, a dynamic adaptive Euclidean distance constraint and The elimination algorithm of RANSAC common constraints can not only ensure the effective elimination of mismatches, but also retain the correct matching points to the greatest extent. The implementation process of this method is as follows figure 2 As shown, the above-mentioned process will be described in detail below.

[0054] 1. Build a dee...

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 matching method and device, and belongs to the technical field of remote sensing image data processing. According to the method, firstly, an improved VGG16 model is adopted to carry out feature extraction, so that extracted feature points have enough abstraction and relatively high positioning precision; then key point screening is carried out on theextracted feature maps to avoid interference of the feature maps with unremarkable features, so that the matching accuracy is improved, and the data processing efficiency is also improved; meanwhile,a dynamic adaptive distance method is adopted to screen to-be-screened matching pairs, automatic configuration is carried out according to data characteristics, high-quality matching pairs can be screened out, and the matching performance of the heterologous remote sensing image is improved. A plurality of groups of different-source remote sensing images are used for testing the method, results show that the method has high adaptability and robustness, the remote sensing image matching method and device are superior to a DELF algorithm in the aspects of adaptability, the number of matching points, distribution, efficiency and the like, and the robust matching performance of the different-source remote sensing images and the remote sensing images with large differences is improved.

Description

technical field [0001] The invention relates to a remote sensing image matching method and device, and belongs to the technical field of remote sensing image data processing. Background technique [0002] With the rapid development of remote sensing technology, multi-type sensors such as visible light, infrared, and synthetic aperture radar (Synthetic Aperture Radar, SAR) have increasingly enriched Earth observation images. The heterogeneous images acquired by different platforms and sensors are complementary to a certain extent, providing a massive data source for in-depth mining of remote sensing information and big data analysis. The matching between images is the core issue of further processing and analysis of heterogeneous images. Due to different imaging mechanisms, wave bands, and time phases, heterogeneous images have huge differences in radiation characteristics and geometric characteristics. The matching between them has always been a difficult research point in ...

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): G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V10/462G06V10/751G06N3/045
Inventor 蓝朝桢施群山张永显卢万杰吕亮崔志祥侯慧太秦剑琪
Owner PLA STRATEGIC SUPPORT FORCE INFORMATION ENG UNIV PLA SSF IEU
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