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

Passenger flow detection method, system and device based on deep learning

A deep learning and detection method technology, applied in the field of passenger flow detection based on deep learning, can solve the problems of misjudgment as a person, large statistical error, and low accuracy rate

Pending Publication Date: 2021-11-02
北京安吉升科技发展有限责任公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] At present, the passenger flow counters used by many enterprises count the passenger flow by identifying the outline and height of the object to judge whether the object is a person, etc. The disadvantages of these methods are that the accuracy rate is not high and the statistical error is large. If the height and outline of an object Similar to a person, it is very likely to be misjudged as a person

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
  • Passenger flow detection method, system and device based on deep learning
  • Passenger flow detection method, system and device based on deep learning
  • Passenger flow detection method, system and device based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0057] see figure 1 , figure 1 Shown is a passenger flow detection method based on deep learning provided by the embodiment of the present application, which aims to identify the outline and height of the object to determine whether the object is a person, so as to count the disadvantages of the passenger flow. This embodiment adopts the method of identifying After judging whether the object is a person by the outline and height of the object, and then monitoring whether the object is moving, so as to achieve the purpose of determining whether it is a person, the specific implementation methods include:

[0058] S101: Collect image information when there are no pedestrians in the environment to be tested as a background sample; collect image information of pedestrians passing by in the environment to be tested at multiple moments as training samples;

[0059] First of all, in order to judge the objects similar to people in the environment to be tested for the first time, the ...

Embodiment 2

[0076] see figure 2 , providing a passenger flow detection system based on deep learning for the embodiment of the present application, including

[0077] The image acquisition module 1 collects image information when there are no pedestrians in the environment to be tested as a background sample; collects image information of pedestrians passing by at multiple moments in the environment to be tested as a training sample;

[0078] Comparison module 2, for performing image subtraction on any training sample and background sample to obtain corresponding feature image samples;

[0079] The preset input module 3 is used to preset human motion image samples;

[0080] The model definition module 4 is used to define a screening model, and the screening model includes a classifier model; importing preset human action image samples and feature image samples into the classifier model to obtain a decision model;

[0081] The video acquisition module 5 is used to collect real-time vide...

Embodiment 3

[0084] see image 3 , to provide a deep learning-based passenger flow detection device for the embodiment of the present application, including the above-mentioned deep learning-based passenger flow detection method, the device includes a connecting piece 7, a universal drive 8, a camera 9, and a camera set on the connecting piece 7 One end of the universal drive 8 is connected to the connector 7, and the other end of the universal drive 8 is connected to the camera 9.

[0085]In some embodiments of the present invention, for the collection of passenger flow information, most of the time, the passenger flow monitoring device needs to perform multi-angle monitoring, so for the device applied to the above-mentioned passenger flow detection method based on deep learning, the universal drive 8 is used, Utilizing the universal drive 8 makes the camera 9 rotate 360 ​​degrees in the vertical plane and the horizontal plane, thereby expanding the monitoring range, and because a fire oc...

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 passenger flow detection method, system and device based on deep learning, and relates to the technical field of passenger flow monitoring. The method comprises: acquiring image information when no pedestrian exists and image information when the pedestrian passes; performing image subtraction on the acquired image to obtain a corresponding feature image sample; importing a preset human body action image sample and the feature image sample into a classifier model to obtain a decision model; collecting real-time video information of a to-be-tested environment, and matching the video information with the test sample based on the decision model; executing the decision model on the test sample; when a result obtained by the feature image samples in the test sample based on the decision model is in a preset first range, performing position comparison on the corresponding feature images, and if the positions are the same, keeping the passenger flow volume value unchanged; and if the positions are different, adding one to the passenger flow volume value. The method can distinguish an object with a height and a contour similar to a person, prevents the object from being misjudged as the person, improves the monitoring precision, and enlarges the monitoring range of the camera.

Description

technical field [0001] The present invention relates to the technical field of passenger flow monitoring, in particular to a method, system and device for passenger flow detection based on deep learning. Background technique [0002] In modern society, the competition among retail enterprises such as shopping malls and chain stores is intensifying, and the core of retail enterprise competition is passenger flow and the purchase conversion rate of passenger flow. How to carry out real-time and dynamic monitoring and statistics on the customer flow of shopping malls, analyze the data and output various analysis results according to the actual operating conditions of enterprises, and provide references for enterprise management decisions have become increasingly concerned issues for many retail enterprises. At the same time, with the development of the economy and society, the retail business model has evolved from traditional passive marketing and extensive marketing to active...

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): G06K9/00G06K9/62
CPCG06F18/214G06F18/241
Inventor 余方敏高尚
Owner 北京安吉升科技发展有限责任公司
Features
  • Generate Ideas
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com