A public area pedestrian flow tracking method based on video data

A technology for public areas and video data, applied in the field of deep learning and big data learning, it can solve the problems of difficult installation and high requirements for camera installation

Inactive Publication Date: 2019-01-18
北京此时此地信息科技有限公司
View PDF6 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the crowd counting of binocular vision has high requirements for the installation of the camera. It is required t

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 public area pedestrian flow tracking method based on video data
  • A public area pedestrian flow tracking method based on video data
  • A public area pedestrian flow tracking method based on video data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0087] Taking an open square as an example, this patent is described as an example.

[0088] Among them, MATLAB is used as the labeling tool, and the clustering algorithm uses the Kmeans algorithm. The regional positioning network can adopt a common network structure, such as the VGG16 network.

[0089] The method includes the following steps in detail:

[0090] Step 1) The obtained video is split into images, and each image is manually marked, and the marked information is the coordinates of the frame in the figure. Among them, the resolution of the video picture should be above 1280×720, and the frame coordinates are divided into horizontal and vertical coordinates of the upper left corner and the lower right corner of the frame.

[0091] Subsequently, the image is up-sampled and down-sampled, and the corresponding border coordinates are multiplied and divided by the corresponding sampling coefficient.

[0092] Finally, the frame is input into the Kmeans algorithm for clu...

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 public area flow tracking method based on video data, which comprises the following steps: obtaining video and video information transmitted from a surveillance camera arranged in a public place, wherein, the video information comprises a surveillance camera number and a shooting time; splitting the video into a series of images in frames; inputting an image into a prediction model trained in advance, and outputting border information, wherein the border information is coordinate information of a border of a human head in the image; comparing the border information ofthe adjacent frames and outputting the coordinates of the position and the trajectory of the human.

Description

technical field [0001] The invention belongs to a method for tracking crowds in public areas based on video data, and relates to deep learning and big data learning. Background technique [0002] With the increasing social material wealth and the improvement of people's living standards, and the construction of various large-scale places, the flow of people is of great significance to industries that depend on the flow of people. [0003] Customers are the carriers of currency and potential buyers of commodities. Studying the flow rules can increase sales opportunities, turn viewers into shoppers, maximize the sales potential of shopping malls, and increase profits. The flow of people is an important measurement tool. Through this accurate quantitative data, not only can the complete running status of public places such as shopping malls, shopping centers, museums or airports be obtained, but also these high-precision data can be used to carry out Effective organizational o...

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): G06K9/00G06T7/20
CPCG06T7/20G06T2207/20081G06T2207/10016G06V20/53
Inventor 李林洋周思远豆泽阳
Owner 北京此时此地信息科技有限公司
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