Research On Intrusion Detection Of Foreign Objects On Outdoor Platform Based On Machine Vision

At present,there is a shielded gate between the waiting area and the rail transit area of the subway platform to ensure the safety of passengers when they are waiting.However,there is a certain gap between the train and the screen door.When the train is running,passengers are caught in the gap,objects are left behind,clipped,and clothes are clamped.There is a large potential safety hazard and accident risk.For the security detection between subway trains and screen doors,the existing methods are mainly artificial lookouts,but this method cannot accurately detect long linear platforms and curvilinear platforms.The automatic detection method mainly uses the laser beam and the infrared light curtain to detect,but the laser beam detection and the infrared light curtain detection are point and area detectors,which may cause high false alarm rate due to vibration.At the same time,for large gap platforms,there is a huge blind area of detection,which does not guarantee vehicle safety.At present,there have been many achievements in the detection of foreign objects in underground platforms.However,there are relatively few foreign objects detected in outdoor subways.Based on this,this paper mainly focuses on the detection of foreign objects in outdoor subways as follows:Extensive access to relevant journal articles and patents for detection of foreign invasion,field surveys and inspections of existing requirements and applications for foreign invasion detection in the subway,reference and refer to sensor performance and detection methods for related obstacle detection,and propose a solution based on machine vision to solve metro outdoor Platform foreign object intrusion detection.Analyze the environmental characteristics of outdoor platforms,define the intrusion of foreign objects,elaborate the characteristics of its intrusion and the detection difficulties,and propose the functional requirements and related performance requirements of the foreign object detection system.For narrow-gap ground stations,this paper proposes a scheme for foreign objects detection based on the ambient light illumination adaptive self-adaptive background.First of all,by designing a strong contrast strip light source for adaptive light,constructing an artificial optical background,and building a camera in the crevice area,shooting In the space gap area,considering the external light changes of the outdoor platform,an adaptive foreign object detection algorithm is designed to achieve foreign object intrusion detection in the space scene.For wide-gap ground stations,this paper proposes a monitoring and inspection method by installing a camera on each fixed door of the screen door.Through the middle of the camera,the whole area to be inspected is monitored,and static texture templates and dynamic backgrounds are constructed.The method of the model is compared to realize the monitoring of the space foreign matter of the wide gap platform.Finally,the simulation experiment environment was built outdoors to verify the feasibility and effectiveness of the designed system and algorithm.