Pedestrian-oriented long-time multi-target tracking method

A multi-target tracking, long-term technology, applied in image data processing, image enhancement, instruments, etc., can solve problems such as algorithm performance degradation

Active Publication Date: 2020-10-30
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF10 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, using the traditional multi-target tracking algorithm in a co

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
  • Pedestrian-oriented long-time multi-target tracking method
  • Pedestrian-oriented long-time multi-target tracking method
  • Pedestrian-oriented long-time multi-target tracking method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0054] like figure 1 , a pedestrian-oriented long-term multi-target tracking method, including the following steps:

[0055] Step 1. Use Socket to connect the system to the remote camera and receive the video data of remote monitoring. The data preprocessing is as follows: every 5 frames of images, the local system receives 1 frame of image, adjusts the image to 224×224, and processes it. The final image image has to be stored in the cache buffer to obtain the video data date;

[0056] Step 2. Read the video data date in the bufier, use the YOLOv3 algorithm to detect the target on the read image, and obtain the position information of all pedestrians involved in the current image, that is, >, Among them, are the coordinates of the upper left corner of the pedestrian detection frame, and store the detection results in a txt file;

[0057] Step 3. Carry out target tracking processing for the image that has completed the detection function;

[0058] Record the coordinates of ...

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 pedestrian-oriented long-time multi-target tracking method. The method belongs to the field of multi-target tracking algorithms. The invention relates to the technical fieldof pedestrian long-time tracking algorithms, in particular to the technical field of pedestrian long-time tracking algorithms, and aims to solve the problem of how to realize correct target association for a long-time shielded target. The method comprises the following steps: data preprocessing, target detection and target tracking, and the target tracking comprises feature extraction, feature measurement, target association, trajectory check and trajectory updating. Preprocessing by data, the receiving of the original data and the processing of the image are realized; subsequently, target detection, pedestrian information related to the currently received image is obtained; and target association is realized for pedestrians involved in a continuous image sequence through target tracking so as to generate a tracking trajectory, and the algorithm realizes correct association of a long-term occlusion target through a secondary comparison mode in consideration of various complex conditions, thereby improving the robustness of the algorithm and improving the accuracy of target association at the same time.

Description

technical field [0001] A pedestrian-oriented long-time multi-target tracking method, the invention belongs to the field of multi-target tracking algorithms, and specifically relates to the technical field of pedestrian long-time tracking algorithms. Background technique [0002] Multi-target tracking is mainly for continuous image sequences, according to the target detection algorithm to obtain the target in the image, and then call the target tracking algorithm to correctly associate the moving target in the continuous image. The multi-target tracking algorithm mainly adopts the technical route of multi-target tracking based on detection. The target tracking part of the multi-target tracking algorithm mainly includes the appearance model and the motion model. The appearance model is mainly to extract the appearance features of the target obtained by the detection algorithm. Estimate the location information of the target in subsequent frames. [0003] The traditional mult...

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): G06T7/246G06N3/04
CPCG06T7/246G06T2207/10016G06T2207/20084G06T2207/30241G06N3/044G06N3/045
Inventor 田玲金琪段贵多罗光春李诗琪高向孚
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