A fast unmanned aerial vehicle image matching method based on fusion of local features

A technology of local features and matching methods, applied in the field of image matching, can solve the problems of large distortion, large changes in viewing angle, and inability to obtain a sufficient number of image homonymic points, and achieve the effect of improving matching accuracy and quantity

Inactive Publication Date: 2018-12-28
GEOVIS CO LTD
View PDF3 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, for multi-view images, the common shortcoming of the existing technology is that it is impossible to obtain a sufficient number of image points with the same name.
Therefore, under the existing feature matching technology conditions,

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
  • A fast unmanned aerial vehicle image matching method based on fusion of local features
  • A fast unmanned aerial vehicle image matching method based on fusion of local features
  • A fast unmanned aerial vehicle image matching method based on fusion of local features

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0045] The present invention will be further described in detail below with reference to the drawings and specific embodiments.

[0046] figure 1 Shown is a fast UAV image matching method based on fusion of local features. The overall steps are:

[0047] Step 1. Perform 3*3 grid segmentation on the reference image and the image to be matched, divide an image into 9 sub-regions, and extract invariant features in the sub-regions; the specific process is:

[0048] a. Perform feature detection on the reference image through the Hessian-Affine affine covariant region detector:

[0049] The initial feature point of Hessian-Affine is determined by the image point x second-order differential Hessian matrix H(L) and its rank DoH;

[0050]

[0051]

[0052] In the formula, L x (x) and L y (x) represents the differential results in the x and y directions after Gaussian smoothing of g(σ) is performed on the target image; with Indicates the second-order differential results in the x and y direct...

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 fast unmanned aerial vehicle image matching method based on fused local features, which comprises the following steps: respectively dividing a reference image and an image tobe matched into 3*3 grid blocks, dividing an image into nine sub-regions, and extracting invariant features in the sub-regions; and extracting the invariant features in the sub-regions; feature descriptors being used to extract feature vectors from invariant feature regions; by comparing the similarities between feature vectors, the initial homonymous feature being distinguished and the stable initial matching being obtained; counting the number of matching points in each grid, and matching the features of MSERs in the region where the number of matching points in the grid is less than the threshold value; the affine invariance of Markov distance being used to delete mismatched point pairs. The invention can provide more stable and more reliable matching points by fusing a plurality of complementary invariant features, so that the matching accuracy and the number are further improved.

Description

technical field [0001] The invention relates to an image matching method, in particular to a fast UAV image matching method based on fusion of local features. Background technique [0002] The main goal of UAV aerial and ground oblique photography data processing is to restore the fine 3D model of ground objects within the area covered by the image. Among them, UAV aerial multi-lens oblique photography can quickly and conveniently obtain multi-view aerial image data and corresponding POS data in a large range, which is an important method for constructing large-area 3D modeling. At the same time, ground oblique photography can flexibly measure and process the ground objects missed and blocked by aerial oblique photography, and it is also an important measurement method in urban 3D modeling. However, the theory and practice of multi-view image matching has always been an important research content of multi-view photography data processing, and it is also the key to realize a...

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): G06K9/62G06K9/46
CPCG06V10/507G06V10/462G06F18/22G06F18/253
Inventor 吴方才范晓敏周馨何晓宁
Owner GEOVIS CO LTD
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