Subway turnstile passing fare evasion identification method based on passenger posture rapid estimation

Abstract

The invention relates to a subway turnstile passing fare evasion identification method based on passenger posture rapid estimation. The method comprises the following steps: 1) acquiring subway turnstile video monitoring information; 2) subway passenger skeleton key point detection; 3) identifying fare evasion behaviors of passengers passing through the gate. Compared with the prior art, on the premise that hardware equipment does not need to be additionally arranged, by designing the subway gate passing fare evasion recognition method for passenger posture rapid estimation, the influence of factors such as the height and the weight of passengers on fare evasion recognition precision can be reduced, and various passing fare evasion behaviors such as crossing, going down and following of a single passenger and a passenger group can be rapidly identified, so gate passing abnormity intelligent alarm and unattended automatic gate channel management can be realized.

Description

technical field [0001] The invention belongs to the technical field of subway intelligent station operation management, relates to the application of video surveillance data in subway stations, and in particular to a method for identifying subway gate pass fare evasion based on rapid estimation of passenger posture. Background technique [0002] With the deepening of rail transit network operation in Shanghai, Beijing and other cities in my country, gates are an important gateway for passengers to enter the payment area. They bear a strong labor load and are closely related to passenger travel safety and efficiency. Fare evasion at gates is not Destruction of public order will also cause huge potential safety hazards. In actual operation scenarios, there are complex and diverse incidents of fare evasion at gates, and fare evasion usually occurs within a short period of time. Therefore, how to improve the accuracy of intelligent recognition of fare evasion at gates And speed is...

AI Technical Summary

Problems solved by technology

[0003] Existing methods can only detect a single passenger’s fare evasion behavior, and seldom consider various fare evasion behaviors such as single passenger or passenger group jumping, drilling and tailgating, and there is still a certain gap between complex and changeable actual application scenarios. In addition, Chinese patent CN201620292646.6 installs a thin detection device on the ground of the gate channel, and uses the principle of weighing to judge whether passengers have fare evasion behavior, but the detection accuracy is easily affected by the weight of passengers; while Chinese patent CN201510144081.7 judges based on image information Whether the human body has fare evasion behavior, but it is still unable to identify the complex and changeable passenger fare evasion behavior. At present, the identification of fare evasion at gates mainly depends on on-site observation, which requires a large number of staff, which increases the operating cost of the subway station, and there are A certain rate of fare evasion and missed detection cannot meet the development requirements of smart subway unmanned stations

Images