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

Group target clustering method based on algebraic graph theory

A technology of algebraic graph theory and group objectives, applied in complex mathematical operations, instruments, character and pattern recognition, etc., can solve problems such as confusion of grouping results, errors in grouping results, and unsuitability for practical application scenarios without prior knowledge

Active Publication Date: 2020-11-13
BEIJING INSTITUTE OF TECHNOLOGYGY
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Early grouping methods include K-means, fuzzy C-means, etc., but these methods need to specify the number of classifications, which are not suitable for practical application scenarios without prior knowledge, and when the state of the group target changes (such as group splitting, merging, etc.), it is easy to cause Clustering results are confusing
Existing clustering technologies usually use a similarity matrix to calculate the clustering results at a certain moment, but the above technologies all have a premise that the similarity between any two targets in the same group is 1, but in practical applications, When the size of the group target is large, the similarity between the targets in the same group may not all be 1, and there may even be a situation where the similarity with other targets in the group is higher, resulting in errors in the grouping results

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
  • Group target clustering method based on algebraic graph theory
  • Group target clustering method based on algebraic graph theory
  • Group target clustering method based on algebraic graph theory

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0032] Such as figure 1 As shown, a group target grouping method based on algebraic graph theory includes the following steps:

[0033] Step 1. It is assumed that the radar performs pulse compression and constant false alarm detection on the echo during a scanning process to obtain a set of effective measurement sets Z in the scanning space at the tth moment t ={z i}={z 1 ,z 2 ,…,z k}, where z i is the state vector of the i-th target, and the state vector includes the position, velocity and acceleration of the target, i=1, 2,..., k, k is the number of targets in the t-th scanning moment;

[0034] Step 2. Calculate the similarity σ between the i-th element and the j-th element in the set Zt ij , wherein, j=1, 2, ..., k; similarity can be defined according to actual needs, and the distance between commonly used targets is used as the discrimination criterion; the present invention adopts the following preferred definitions:

[0035]

[0036] Among them, γ is the preset...

Embodiment 2

[0057] In this embodiment, taking the group target as a flock of birds as an example, a method for grouping group targets based on algebraic graph theory is described in detail.

[0058] Step 1. Use a high-resolution radar containing only distance information to observe multiple flocks of birds; high-resolution radar bandwidth B=785MHz, distance resolution △=c / 2B=0.19m, where c is the speed of light; for the data after a certain scan After performing pulse compression and constant false alarm detection, a total of 6 effective echoes were detected, and the distances to the radar were 300m, 315m, 320m, 298m, 305m, and 318m; mark the targets as No. 1 to No. 6 in the above order;

[0059] Step 2. Calculate the similarity between objects. According to the bird habits and flight characteristics, the threshold is set to 6m; when the distance between two birds is less than 6m, it is judged to be similar; when the distance between two birds is greater than 6m, it is judged to be dissim...

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 provides a group target clustering method based on an algebraic graph theory. Constructing a connected graph in the monitoring space by utilizing the algebraic graph theory and the relation information between the measurements; real-time grouping of the group targets is completed by searching the connected graph in the monitoring space, the limitation condition that the targets in the same group are required to be completely similar in a traditional method is broadened, and grouping errors caused by the fact that the similarity of the targets in the group is not completely 1 canbe avoided; in addition, the data dimension is not changed when the similarity and the clustering matrix are calculated, that is, the method can be used for clustering processing of the group targetson the premise of not changing an existing radar data processing structure.

Description

technical field [0001] The invention belongs to the field of information processing, in particular to a method for grouping objects based on algebraic graph theory. Background technique [0002] Grouping detection is a unique data processing step of group target tracking, and it is also one of the important signs different from ordinary multi-target tracking. The basic idea of ​​group detection is to divide all valid measurements obtained by a certain scan of the sensor into several subsets according to certain criteria. In order to accurately start the track of group targets in the future, group detection must be able to accurately reflect the distribution characteristics of different group targets in the measurement space, and reflect the group dynamic change process in real time. Early grouping methods include K-means, fuzzy C-means, etc., but these methods need to specify the number of classifications, which are not suitable for practical application scenarios without p...

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): G06K9/00G06F17/16
CPCG06F17/16G06V20/10Y02A90/10
Inventor 王锐周超姜琦胡程
Owner BEIJING INSTITUTE OF TECHNOLOGYGY
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