Pedestrian relationship identification method, device and system and electronic device

A technology of relationship recognition and recognition algorithm, which is applied in the field of image processing, can solve problems such as difficult to apply, unable to identify people, etc., and achieve a strong practical effect

Active Publication Date: 2019-04-23
恒睿(重庆)人工智能技术研究院有限公司
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Benefits of technology

This technology helps improve how people interact with their environment by providing an accurate way for them to recognize relationships based on factors like distance from others or other things around us. It uses algorithms to analyze where individuals meet at specific locations over long periods of time (usually hours) to create a map showing which groups they belong together most likely to form. By comparing these maps against predetermined thresholds, it becomes possible to predict who will come next after another person's meeting. Overall this makes communication easier even when there aren’t many direct neighbors involved.

Problems solved by technology

The technical problem addressed by this patented system relates to accurately recognizing how well children have different ways of being together when walking around their own home without getting lost due to lack of communication during meetings at restaurants. This helps companies better manage risk management strategies like providing more effective advertising campaigns that target specific areas where potential targets may be located.

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 relationship identification method, device and system and electronic device
  • Pedestrian relationship identification method, device and system and electronic device
  • Pedestrian relationship identification method, device and system and electronic device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0032] Embodiment 1 of the present invention provides a pedestrian relationship identification method, see figure 1 The flow chart of the pedestrian relationship recognition method shown in the figure, the method includes the following steps:

[0033] Step S102, analyzing the orientation information of pedestrians within the target range in real time according to the group recognition algorithm to obtain the target group.

[0034] The group recognition algorithm can identify groups by tracking the changes in the degree of intimacy between pedestrians and groups, and judging the group belonging of pedestrians. Using the orientation information of each pedestrian, identify the group belonging of each pedestrian through a certain group identification algorithm. The target group can be any of a number of groups. The target range can be set according to the needs.

[0035] Step S104, extracting facial features of each group member in the target group.

[0036] see ima

Embodiment 2

[0061] Embodiment 2 of the present invention provides a pedestrian relationship recognition device, see Figure 6 The structural block diagram of the pedestrian relationship recognition device shown, the device includes:

[0062] The group determination module 61 is used to analyze the orientation information of pedestrians in the target range in real time according to the group recognition algorithm to obtain the target group; the feature extraction module 62 is used to extract the face features of each group member in the target group; the scoring module 63, It is used to perform two-by-two cross-validation on the group members according to the face features to obtain the kinship score between each two group members; the recognition module 64 is used to determine the target group member according to the kinship score and the preset threshold, and according to Target group members and kinship scores generate a kinship network; in the kinship network, target group members ar

Embodiment 3

[0066] Embodiment 3 of the present invention provides a pedestrian relationship recognition system, see Figure 5 The schematic block diagram of the structure of the pedestrian relationship recognition system shown, the system includes:

[0067] Family library, image acquisition module, group library, group identification module, face detection module and relative identification module; image acquisition module is used to collect pedestrian information and send pedestrian information to group library; group library is used to receive and store pedestrian information ; The group database is connected to the group recognition module by communication; the group recognition module is used to analyze the pedestrian information within the target range in real time according to the group recognition algorithm to obtain the target group; the face detection module is used to extract the faces of each group member in the target group features; the face detection module communicates wi

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 pedestrian relationship identification method, device and system and an electronic device, and relates to the technical field of image processing. The method comprises the following steps of analyzing azimuth information of pedestrians in a target range in real time according to a group identification algorithm to obtain a target group; extracting face features of each group member in the target group; performing pairwise cross validation on the group members according to the face features to obtain a relative relationship score between every two group members; determining a target group member according to the relative relationship score and a preset threshold value, and generating a relative relationship network according to the target group member and the relative relationship score; connecting the target group members in the relative relationship network through edges, and taking the relative relationship score as the weight. According to the method, the complex relative relation chain which may appear in the pedestrian group can be identified, and the practicability is stronger.

Description

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

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
Owner 恒睿(重庆)人工智能技术研究院有限公司
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