AI-powered crowd analysis and intelligent guidance device

TWM687156UActive Publication Date: 2026-09-01TAIPEI RAPID TRANSIT CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
TW115205193
Authority / Receiving Office
TW · TW
Patent Type
Utility models
Current Assignee / Owner
Filing Date
2026-06-05
Publication Date
2026-09-01
Estimated Expiration
2036-06-04

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

An AI-powered crowd flow analysis and intelligent guidance device comprises a data acquisition module, an image processing host, a correction and calculation module, a diversion indication module, a visualization display module, a guidance output module, and a monitoring display module. This device can provide real-time overflow guidance and use a threshold-triggered diversion method to allow passengers to disperse naturally, thereby reducing reliance on manual commands and the workload of on-site personnel, and effectively reducing bottleneck congestion and pushing risks. It also provides a mechanism for estimating and displaying channel congestion levels that are updated in a short time, combining image recognition and ticket data correction to improve judgment accuracy and reduce the impact of data delays on scheduling decisions. Furthermore, its modular design allows for integration with existing operating systems, facilitating cross-station deployment and expansion, further enhancing system replicability, scalability, and overall operational efficiency.
Need to check novelty before this filing date? Find Prior Art

Claims

1. An AI-powered pedestrian flow analysis and intelligent guidance device, comprising: A data acquisition module is used to acquire image data and ticket data from at least one channel; an image processing host is connected to the data acquisition module to receive the image data and ticket data transmitted by the data acquisition module. The image processing host is equipped with an image recognition module and an edge computing node. The image recognition module is used to identify passenger flow from the image data to output an estimated value of the number of people or the flow of passengers entering the channel. The edge computing node is used to perform image recognition and classification calculation in real time to reduce latency and improve resilience; a correction calculation module is connected to the image processing host to perform regression correction on the estimated value of the image recognition result based on the ticket data to generate a correction coefficient or a corrected passenger flow value; a diversion indication module is connected to the correction calculation module to convert the corrected passenger flow value into a congestion level of at least level two according to a preset threshold, and present diversion guidance information accordingly. A visualization display module, connected to the diversion indicator module, is used to display the congestion level and / or guidance information above the passage or in the entrance area to present the passage status in real time; A guidance output module, connected to the diversion indicator module, is used to display the congestion level and / or guidance information of each car or section on the platform level to provide diversion instructions for the passage or waiting area. It also includes a monitoring and display module connected to the image processing host, the correction calculation module and the diversion indicator module, to integrate the calculation results of each module and display the congestion status, historical trends and alarms of each channel through the operation interface to support scheduling decisions, and adjust the policy threshold and resource configuration as needed.

2. The AI-powered pedestrian flow analysis and intelligent guidance device as described in claim 1, wherein, The data acquisition module is a CCTV (Closed-Circuit Television) camera.

3. The AI-powered crowd analysis and intelligent guidance device as described in claim 1, wherein, The data acquisition module further connects to the gate host and the ticket host installed on the gate to obtain the ticket data and gate status data.

4. The AI-powered crowd analysis and intelligent guidance device as described in claim 1, wherein, This ticket information is for ticket entry and exit data.

5. The AI-powered pedestrian flow analysis and intelligent guidance device as described in claim 1, wherein, The image recognition module is used to calculate the number of people or traffic entering the channel every 10 seconds to generate the estimated value.

6. The AI-powered pedestrian flow analysis and intelligent guidance device as described in claim 1, wherein, This diversion indicator module displays an arrow pointing to an alternative channel to provide diversion guidance information.

7. The AI-powered pedestrian flow analysis and intelligent guidance device as described in claim 1, wherein, The diversion indicator module transmits the congestion level and / or guidance information to the visualization display module, the guidance output module, and the monitoring display module via a web application interface (Web API).

8. The AI-powered pedestrian flow analysis and intelligent guidance device as described in claim 1, wherein, The visualization module displays the congestion level of the passage through congestion indicator lights, which use red, yellow, and green human-shaped prompts to indicate the three levels of congestion.

9. The AI-powered pedestrian flow analysis and intelligent guidance device as described in claim 8, wherein, When the congestion level of a certain passage reaches a preset threshold, the indicator light for that passage turns red and the passage indication is reduced, while the indicator light for the alternative passage turns green and displays a human-shaped indicator light to facilitate overflow guidance.

10. The AI-powered crowd analysis and intelligent guidance device as described in claim 1, wherein, The guidance output module includes one or a combination of guidance light strips, ground guidance signs, application (App) push notifications, or in-site screen information.

11. The AI-powered pedestrian flow analysis and intelligent guidance device as described in claim 1, wherein, The monitoring module is a management dashboard.