Method and system for detecting abnormal events at railway crossings based on generative confrontation network

Abstract

The invention discloses a method and system for detecting abnormal events at railway crossings based on generative confrontation networks, including: collecting video of railway crossings to be detected; processing the video of railway crossings to be detected, judging whether detection of abnormal events is required, , the abnormal event detection is suspended, and if the railway crossing railing is lowered, the next step is entered; based on the trained generative confrontation network, the abnormal event detection is performed on the railway crossing video to be detected, and the abnormal event detection result is output. Based on the idea of ​​generative confrontation network, an anomaly detection network architecture is proposed, which learns the appearance and motion features in normal videos, and realizes anomaly detection with few or no negative samples. Using a dual-stream architecture and adding a memory enhancement module improves the network's ability to acquire motion information, inhibits the network's ability to reconstruct abnormal events, increases the reconstruction error, and further improves the network's detection effect on abnormal events.

Description

technical field [0001] The invention relates to the technical field of railway safety detection, in particular to a method and system for detecting abnormal events at railway crossings based on a generative confrontation network. Background technique [0002] The statements in this section merely provide background related to the present disclosure and do not necessarily constitute prior art. [0003] The purpose of abnormal event detection is to detect uncommon, abnormal objects or behaviors in videos or images that may bring hidden dangers to the safety of production and operation. [0004] Railway crossing abnormal event detection is to monitor the railway crossing in real time, detect and locate abnormal events in the monitoring area, such as pedestrians stranded, landslides, etc. Through the real-time detection of the monitoring area, the factors that affect the safe operation of the train can be effectively discovered, and accurate information can be provided to the m...

AI Technical Summary

Problems solved by technology

Most of the unguarded railway crossings just automatically drop the railings, but they cannot identify whether there are foreign objects or pedestrians stranded in the railway crossings.

Therefore, at present, most unmanned railway crossings still need a lot of manpower to be on duty before monitoring. Considering that the on-duty personnel usually cannot concentrate for a long time, it is difficult to achieve the desired effect

The data shows that there are still many railway crossings, which will consume a lot of manpower and material resources, and the safety is not high

[0006] At present, a series of anomaly detection methods for railway crossings have been proposed, but most of them are hardware-based methods with high cost and low intelligence

Common algorithms based on video images are greatly disturbed by factors such as light and shadow, and cannot guarantee the accuracy of detection in complex backgrounds, and there are problems such as waste of computing resources, and many types of abnormal events make it difficult to make data sets.

Images