A Road Network Trajectory Clustering Analysis Method Based on Improved DPC Algorithm

A road network and trajectory clustering technology, which is applied in the field of cluster analysis, can solve the problems that the classic trajectory cannot represent the whole situation, and there are too many parameter settings, so as to reduce the influence of custom parameters, improve adaptability, and get rid of human interference.

Active Publication Date: 2021-09-28
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the existing road trajectory clustering cannot describe the hotspots well, and the obtained classic trajectory often cannot represent the overall situation.
In addition, the traditional DPC algorithm is subject to too many parameter settings of human intervention, and it is not a closed system and method

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 Road Network Trajectory Clustering Analysis Method Based on Improved DPC Algorithm
  • A Road Network Trajectory Clustering Analysis Method Based on Improved DPC Algorithm
  • A Road Network Trajectory Clustering Analysis Method Based on Improved DPC Algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0072] A kind of road network locus cluster analysis method based on improved DPC algorithm that the preferred embodiment of the present invention provides, comprises the following steps of carrying out successively:

[0073] A road network track clustering analysis method based on the improved DPC algorithm, comprising the following steps carried out in sequence:

[0074] S1: Data acquisition: Use vehicle-mounted and ground-based road track recording equipment to collect track data of moving objects, or collect GPS data of different moving objects, and use track data or GPS data as input data:

[0075] S2: Trajectory movement expression: obtain the sub-trajectory sequence connected by feature points from the input data of step S1;

[0076] Extract valid global trajectory data from the input data in step S1; build a model through linear interpolation and semantic extension based on the local interpolation model, output the trajectory file, and select a class of anomalous point...

Embodiment 2

[0081] In this embodiment, on the basis of Embodiment 1, the expression of the trajectory movement in step S2 has the following steps:

[0082] S21: Establish a trajectory model based on the line segment trajectory representation, extract the stay point from the input data in step S1 for semantic expansion, and convert the stay point trajectory into a position trajectory;

[0083] S22: Road trajectory data expression based on local interpolation model: After discretizing the road trajectory grid, use the inverse distance weight method to calculate the attribute values ​​of the trajectory segments falling in each grid, and calculate the trajectory segment according to the attributes of the adjacent sampling points perform interpolation;

[0084] S23: Find the feature points based on the separation method of the trajectory segment of the angle size, and obtain the sub-trajectory sequence connected by the feature points.

Embodiment 3

[0086] In this embodiment, on the basis of Embodiment 1, the aggregation distance CD in the step S3 to calculate the distance between sub-trajectories has the following steps:

[0087] S31: For the sub-trajectory segments formed after any two trajectories are expressed, calculate the aggregation distance CD between the sub-trajectory segments, assuming two sub-trajectory segments ST i =b i e i and ST j =b j e j , using d ⊥ (ST i , ST j ) represents the vertical distance between two sub-trajectories, using d || (ST i , ST j ) represents the parallel distance between two sub-trajectories, using d θ (ST i , ST j ) represents the angular distance between two sub-trajectories;

[0088] S32: Sub-track ST i and ST j The aggregation distance CD(ST i , ST j ), that is, the similarity between trajectories, expressed as the weighted average of three normalized distances;

[0089] S33: Calculate the aggregation distance CD between any two sub-trajectories, and finally ob...

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 road network trajectory cluster analysis method based on an improved DPC algorithm, which belongs to the field of cluster analysis methods. The method includes: S1, effective collection of road moving object trajectory; Aggregate distance CD to calculate the distance between sub-trajectories, which combines vertical distance, parallel distance and angular distance, and obtains the similarity measure between trajectories based on CD distance; S4. According to the fusion CD distance matrix, use the improved DPC algorithm Carry out sub-trajectory clustering; S5. For the obtained clustering result, extract the classic trajectory of the clustering result as a new behavior pattern of the entire road network. The present invention improves the accuracy of road trajectory measurement, reduces the influence of user-defined parameters compared with traditional density clustering algorithms, improves the interaction with users, and helps to find meaningful trajectories in road networks model.

Description

technical field [0001] The invention belongs to the field of cluster analysis, and relates to a road network track cluster analysis method based on an improved DPC algorithm. Background technique [0002] The full name of the clustering algorithm based on density peaks is the clustering algorithm based on fast search and find of density peaks (English name: clustering by fast search and find of density peaks, abbreviated as DPC). It is a clustering algorithm proposed in Science in 2014, which can automatically find cluster centers and realize efficient clustering of arbitrary shape data. [0003] With the rapid development of technologies such as smart terminals, mobile positioning, and wireless communications, a large amount of trajectory data constrained by road networks has been collected in application fields such as transportation and logistics. Using trajectory data to analyze paths can reflect the movement and behavior patterns of moving objects. However, the existi...

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 Patents(China)
IPC IPC(8): G06K9/62G06F16/29
CPCG06F16/29G06F18/2321
Inventor 牛新征刘鹏飞郑云红望馨刘翔宇
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA
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