Multi-attribute coding and dynamic weight-based graph matching method

A technology of dynamic weights and matching methods, applied in image coding, image analysis, image data processing, etc., can solve the problems of limiting the application scope of the algorithm, being stretched, the scope of application, and the lack of robustness.

Active Publication Date: 2018-10-16
ZHEJIANG UNIV
View PDF10 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although this method is efficient and stable, it is only for the calculation of the similarity of polygonal graphics, and the graphics need to be preprocessed so that the number of sides of the two graphics must be equal. Various conditions limit the scope of application of the algorithm
A method and system for constructing and matching graphics in a Chinese patent (patent application publication number: CN105335444A, authors: Liu Lu, Wang Xinghua, Lv Xiaoqing) encodes the properties of linear graphics and non-linear graphics respectively to realize complex structures, The geometric features of various types of primitives are expressed and matched, but the graphic encoding method used in this patent cannot distinguish mirror-symmetrical graphics
[0005] The above methods have certain deficiencies in the scope of application, robustness, etc., which make them somewhat stretched in the process of engineering application.

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
  • Multi-attribute coding and dynamic weight-based graph matching method
  • Multi-attribute coding and dynamic weight-based graph matching method
  • Multi-attribute coding and dynamic weight-based graph matching method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0081] In this embodiment, according to the pattern matching method of the present invention, figure 2 Calculate the similarity distance between the two graphs and find the corresponding edge.

[0082] The pattern matching process of the present invention is as follows figure 1 The specific implementation process is as follows:

[0083] (1) Input figure 2 The two graphs of the two graph objects to be compared are shown as figure 2 (a) and figure 2 (b), and transform it into an edge-attributed adjacency graph G respectively 1 , G 2 , that is, the geometric type and relative length information of the edge of the graph are stored in the vertex set V corresponding to the adjacency graph 1 and V 2 At the same time, the adjacency information between graph edges and edges is stored in the edge set E corresponding to the adjacency graph 1 and E 2 , the conversion result of Figure 2(a) is as follows image 3 shown.

[0084] (2) According to G 1 , G 2 The information of...

Embodiment 2

[0123] In this embodiment, given a source graphic, the most similar graphic is retrieved from the graphic library.

[0124] source graphics such as Image 6 shown, the graphics library such as Figure 7 shown. According to the graphic matching method of the present invention, the similarity distance between the source graphic and each graphic in the graphic library is calculated respectively, and the results are shown in Table 5.

[0125] Table 5 The similarity distance between each graph in the graph library and the source graph

[0126] graphics

[0127] Table 6 Corresponding edges between the source graph and graph (5) in the graph library

[0128] source graphics

[0129] It can be seen from the calculation results in Table 5 that the best matching graph of the source graph is Figure 7The figure (5) in , conforms to the human visual perception law, and the corresponding edge calculation results in Table 6 can also illustrate that the figure matchin...

Embodiment 3

[0131] This embodiment is compared with the matching algorithm in a Chinese patent for a method for constructing a graph, a matching method and a system (Patent Application Publication No.: CN105335444A, authors: Liu Lu, Wang Xinghua, Lv Xiaoqing), which shows that the matching algorithm of the present invention can Mirror symmetrical figures are distinguished. Calculated by two matching algorithms Figure 8 The similarity distance between Figure (1), Figure (2), Figure (3), where Figure (2) and Figure (3) are mirror-symmetrical, and the calculation results are shown in Table 7.

[0132] Table 7 The similarity distance calculation results of the two matching algorithms

[0133] similarity distance

[0134] As can be seen from the calculation results in Table 7, due to the use of the current edge as the benchmark, the angle coding method that defines the included angle in a clockwise direction, and the method of hierarchically calculating the graph edge adjacency att...

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 multi-attribute coding and dynamic weight-based graph matching method. The method comprises the steps of inputting two to-be-compared graph objects and converting the graph objects into edge attribute adjacency graphs; according to information of vertexes and edges in the edge attribute adjacency graphs, performing multi-attribute coding including basic attributes and adjacency attributes for the vertexes in the edge attribute adjacency graphs; hierarchically calculating the similarity between every two vertexes in the edge attribute adjacency graphs by adopting a dynamic weight method, and performing storage to form a vertex similarity mapping matrix; and when a weighted sum of maximum weight matching of the vertex similarity mapping matrix is maximum, obtaininga similarity distance between the two graphs and corresponding edges. The basic attributes and the adjacency attributes of the edges of the graphs are comprehensively considered, and the coding and the similarity calculation are performed according to the respective characteristics, so that the stability and reliability of graph matching are improved and the application range is expanded.

Description

technical field [0001] The invention belongs to the field of graphic image processing, in particular to a graphic matching method based on multi-attribute coding and dynamic weight. Background technique [0002] 3D registration refers to obtaining the accurate 3D coordinates and poses of objects in 3D space through computer graphics analysis, and binding and splicing virtual objects generated by computer into the real 3D space according to the obtained 3D coordinates to achieve the real environment and virtual objects. Accurate and seamless integration of the augmented reality system is the basis for realizing various augmented reality systems, and it is also one of the core problems to be solved. [0003] The 3D registration method based on model keyframes needs to extract the geometric figures in the monocular image in the online stage, and match the most similar keyframe figures from the figures stored in the offline stage. However, due to the discrete processing of the ...

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): G06T7/30G06T9/20
CPCG06T7/30G06T9/20
Inventor 刘振宇王科刘达新谭建荣
Owner ZHEJIANG 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