Escalator passenger falling detection algorithm based on deep learning

An escalator and deep learning technology, applied in computing, computer components, instruments, etc., can solve problems such as high labor costs, slack monitoring staff, and inability to deal with safety accidents in a timely manner

Active Publication Date: 2018-11-13
SOUTH CHINA UNIV OF TECH
View PDF4 Cites 37 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] With the continuous advancement of my country's modernization process and the continuous improvement of people's economic living standards, more and more public facilities appear in public places to facilitate people's daily production and life. As a public facility that facilitates passengers to travel, escalators are used in shopping malls. It can be seen everywhere in public places such as office buildings, subway stations, etc. However, the convenience of travel has caused a series of safety problems, such as passenger congestion on the escalato

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
  • Escalator passenger falling detection algorithm based on deep learning
  • Escalator passenger falling detection algorithm based on deep learning
  • Escalator passenger falling detection algorithm based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0121] The present invention will be further described below in conjunction with specific examples.

[0122] The escalator passenger fall detection algorithm based on deep learning provided by this embodiment first uses the FHOG descriptor and the SVM classifier to detect the passenger's face, uses KCF to track the passenger's face, and creates a passenger trajectory list based on the passenger's face information. Then use transfer learning to retrain the yolo2 algorithm model to detect the passenger's body, match the passenger's face and passenger's body, add the personal information to the trajectory list, and then use the openpose deep learning algorithm to extract the passenger's bone joint point sequence, and match the passenger's body with the passenger's skeleton Joint point sequence, adding the bone joint point information to the track list, and finally analyzing the bone joint point information in the track list to detect the passenger's fall behavior. The algorithm fl...

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 an escalator passenger falling detection algorithm based on deep learning. The method comprises the following steps: 1) collecting a video image that a passenger takes an escalator; 2) detecting a passenger face by using a FHOG describer and a SVM classifier; 3) tracking the passenger face by using KCF, and newly establishing a passenger trajectory list based on the passenger face information; 4) retraining yolo2 algorithm model to detect the passenger body by using migration learning; 5) matching the passenger face and the passenger body, and adding the body information in the trajectory list; 6) extracting a passenger bone joint sequence by using an openpose deep learning algorithm; 7) matching the passenger body with the passenger bone joint sequence, and addingthe bone joint information in the trajectory list; and 8) analyzing the bone joint information in the trajectory list and detecting the passenger falling behavior. Through the algorithm disclosed by the invention, the falling behavior of the passenger taking the escalator can be detected, an emergency scheme can be timely started when discovering the falling behavior, so that the safety hazard canbe reduced to a minimum.

Description

technical field [0001] The present invention relates to the technical field of image processing and behavior recognition, in particular to a deep learning-based detection algorithm for passengers falling down in an escalator. Background technique [0002] With the continuous advancement of my country's modernization process and the continuous improvement of people's economic living standards, more and more public facilities appear in public places to facilitate people's daily production and life. As a public facility that facilitates passengers to travel, escalators are used in shopping malls. It can be seen everywhere in public places such as office buildings, subway stations, etc. However, the convenience of travel has caused a series of safety problems, such as passenger congestion on the escalator, passengers going backwards, passengers running, passengers falling, etc., these Behaviors, especially passenger falls, will cause serious safety accidents. It is necessary to mo...

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): G06K9/00G06K9/62G06K9/66
CPCG06V40/23G06V40/172G06V30/194G06F18/2411
Inventor 田联房吴啟超杜启亮
Owner SOUTH CHINA UNIV OF TECH
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