Subway door foreign matter detection method

Abstract

The invention discloses a subway door foreign matter detection method. The method comprises the following steps: S1, establishing a deep learning model; S2, in the closed state of the train doors, shooting a reference image when no foreign matter exists between each train door and the shielding door of the subway platform; S3, shooting real-time images between each train door and the shielding door of the subway platform in a train door closing state when the subway runs; S4, performing similarity comparison on the real-time image of each train door and the corresponding reference image to obtain a similarity value; S5, judging whether the similarity value is greater than or equal to a set threshold value, if so, judging that no foreign matter exists, turning to S8, and if not, turning to the next step; S6, carrying out judgment by adopting a deep learning model, and obtaining and displaying a judgment result; S7, performing corresponding processing according to a judgment result; and S8, ending. Through combination of similarity calculation and the deep learning model, real-time detection of the subway foreign matter is realized, and the foreign matter detection efficiency and accuracy are improved.

Description

technical field [0001] The invention relates to the technical field of subways, in particular to a method for detecting foreign objects in subway doors. Background technique [0002] At present, the subway has become an important means of transportation for people to travel. The safety of the subway is related to the safety of thousands of people and the happiness of countless families. The detection of subway obstacles is a top priority for the safe operation of the subway. [0003] At present, there are many methods for detecting subway obstacles, including infrared light curtain method and laser detection method. The infrared light curtain method is composed of an infrared transmitter, an infrared receiver, and a host. Since the signal emitted by the transmitter is infrared light, the spot increases gradually with the increase of the detection distance, so it is not suitable for long-distance detection. The laser control method is similar to the infrared light curtain me...

AI Technical Summary

Problems solved by technology

The laser control method is similar to the infrared light curtain method. For curved platforms, after the number of devices increases, the false alarm rate increases accordingly. The devices are installed between the train and the platform door, and the detection area is limited. Although it does not invade the train limit, it exceeds the platform. equipment limits, there are also great hidden dangers to driving safety

[0004] At the same time, due to the long distance and dark light between the subway door and the platform door, the high frequency of train departures, and the limited operating time of the driver, the current manual detection has a large number of false detections.

Images