Station passenger help calling service method based on image recognition

Abstract

The embodiment of the invention provides a station passenger help calling service method, device and equipment based on image recognition and a computer readable storage medium. The method comprises the following steps: acquiring a platform video image; performing identification processing on the video image through a preset target detection algorithm to obtain a two-channel picture; inputting thetwo-channel picture into a pre-trained behavior recognition model, and determining an abnormal behavior category corresponding to the two-channel picture; according to the abnormal behavior category,sending the video image corresponding to the abnormal behavior category to a comprehensive control room display screen, a cloud and / or a platform worker handheld terminal. In this way, real-time detection and abnormal information pushing of the abnormal condition of the platform are achieved, and the working intensity of platform workers is reduced.

Description

technical field [0001] Embodiments of the present disclosure generally relate to the rail transit field, and more specifically, relate to an image recognition-based station passenger recruitment service method, device, equipment and computer-readable storage medium. Background technique [0002] The construction and operation of my country's urban rail transit is entering a period of large-scale and networked operation. However, with the continuous expansion of the scale of domestic rail transit operations and the sharp increase in passenger flow, the operational safety of urban rail transit systems is facing new challenges. [0003] At present, the research on rail transit passenger flow recognition is in its infancy, and the judgment of safety status is still in the manual monitoring mode, mainly relying on the staff to observe the monitoring images provided by the cameras in specific areas, and there is no active perception ability for passenger behavior recognition. At ...

AI Technical Summary

Problems solved by technology

[0003] At present, the research on rail transit passenger flow recognition is in its infancy, and the judgment of safety status is still in the manual monitoring mode, mainly relying on the staff to observe the monitoring images provided by the cameras in specific areas, and there is no active perception ability for passenger behavior recognition

At the same time, due to the limitations of the camera, it is difficult to monitor passengers in all directions

Images