Passenger intention recognition method and system based on elevator scene-behavior association

By introducing object detection and context injection methods into the elevator system, and utilizing the I3D network and the spatiotemporal vision Transformer model, the problem of traditional elevator systems being unable to respond to passenger needs in real time is solved. This enables accurate identification of passenger intentions and personalized responses, thereby improving the operational efficiency and safety of the elevator system.

CN120220010BActive Publication Date: 2026-06-12SUZHOU UNIV OF SCI & TECH
0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SUZHOU UNIV OF SCI & TECH
Filing Date
2025-02-10
Publication Date
2026-06-12

AI Technical Summary

⚠Technical Problem

Traditional elevator management systems cannot respond to passengers' dynamic needs in real time, resulting in long waiting times, overcrowding, and safety hazards. They also lack accurate identification of passengers' intentions and are particularly inadequate in response to emergencies.

⚗Method used

A passenger intent recognition method based on elevator scene-behavior association is adopted. Through target detection, local feature extraction, motion trajectory aggregation and context injection, the I3D network and spatiotemporal vision Transformer model are used to capture passenger behavior and intent, and enhance the association between scene and behavior subject.

🎯Benefits of technology

This improved the accuracy of the elevator system in recognizing passenger intentions, enabled personalized responses, and enhanced the response efficiency and safety of the elevator system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The application discloses a passenger intention recognition method and system based on elevator scene-behavior association, and belongs to the field of computer vision. The method first generates a detection frame of an object by using a target detection module; a local feature extraction module is introduced to obtain a feature map of the target detection frame area as a region of interest, so as to obtain position information and appearance features of each pedestrian in each frame; a motion trajectory aggregation module is introduced to integrate the aggregated features of each object extracted in the video sequence across time, so as to capture and construct the motion trajectory of each object over time, and infer different motion trajectories along the time dimension; and the original video frame is taken as context information and injected into the interaction feature, so as to enhance the association between the scene and the behavior subject. The application can better adapt to the recognition of passenger intention in the elevator scene, so as to more accurately analyze whether the passenger has the demand to take the elevator, and realize a more intelligent elevator system.
Need to check novelty before this filing date? Find Prior Art