Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
EDGE END-ORIENTED PEDESTRIAN TRACKING AND ACCURATE PEOPLE COUNTING METHOD
Document Type and Number:
WIPO Patent Application WO/2021/147826
Kind Code:
A1
Abstract:
An edge end-oriented pedestrian tracking and accurate people counting method, which relates to the field of pedestrian tracking and people counting on the basis of an artificial intelligence component in an edge computing scenario. The method comprises: initially establishing and maintaining in real time a pedestrian dynamic tracking list; acquiring pedestrian result information from an an edge end artificial intelligence component, and establishing a pedestrian static detection list on the basis of same; and accurately matching the pedestrian static detection list with the pedestrian dynamic tracking list. The costs and performance limitations of the edge end cause edge end computing capability to be limited, leading to the failure of accurate pedestrian matching. At which time, in the present method, intelligent fuzzy matching is performed on a pedestrian according to pedestrian dynamic tracking information. A pedestrian dynamic tracking list is updated in real time according to a matching result, the number of pedestrians is accurately counted within a statistical area, and if the accurate counting fails, a delayed counting strategy is used for the pedestrian in the method.

Inventors:
HUANG ZHANGQIN (CN)
SHENG MENGXUE (CN)
ZHANG SHUO (CN)
LI HONGLIANG (CN)
Application Number:
PCT/CN2021/072538
Publication Date:
July 29, 2021
Filing Date:
January 18, 2021
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
UNIV BEIJING TECHNOLOGY (CN)
International Classes:
G06T7/277
Foreign References:
CN111260696A2020-06-09
CN109754409A2019-05-14
CN103049787A2013-04-17
CN110443210A2019-11-12
CN106778570A2017-05-31
CN109255802A2019-01-22
US20150310275A12015-10-29
Other References:
ZHANG, YAN: "Research on Pedestian Detection and Tracking Technology in Indoor Scene", CHINESE MASTER'S THESES FULL-TEXT DATABASE, 1 May 2017 (2017-05-01), pages 1 - 78, XP055831183, ISSN: 1674-0246
Attorney, Agent or Firm:
BEIJING SIHAI TIANDA INTELLECTUAL PROPERTY AGENCY LTD. (CN)
Download PDF: