Machine vision-based foreign object monitoring device between rail transit platform doors and trains

Abstract

The present invention is a device for monitoring foreign objects between rail transit platform doors and trains based on machine vision. The device collects visible light images, infrared light images and depth images between the rail transit platform doors and trains, and integrates three-dimensional images in a three-dimensional space, which can cover the shooting angle of view. All objects in the image can be identified by analyzing the three-dimensional space image data, including whether there are foreign objects in the track area, whether the train is in the station, the number of trains, the switch status of the platform door and the train door, and the door gap of the platform sliding door. , the door seam of the train door, whether there is foreign body intrusion between the train and the platform door, the size, type and whether the foreign body affects the driving safety and other major events, drive the foreign body alarm safety relay, and transmit the monitoring results and live images in real time. It can realize no dead angle in the area, all-area detection, not affected by vehicle type, clearance space size, vibration and environmental illumination, suitable for underground, ground and above-ground platform environments, and suitable for unmanned environments.

Description

technical field [0001] The invention belongs to the field of rail transit safety assurance, and relates to a machine vision-based foreign object monitoring device between a rail transit platform door and a train. Background technique [0002] At present, rail transit such as subways, light rails, bullet trains, and high-speed rails has generally or begun to install platform doors. However, some incidents that affect safety often occur in the gap (risk space) between the platform doors and trains, which has caused many fatalities and major delays. The risk space refers to the three-dimensional area between the platform door and the train as the front and rear, and the plane where the platform pedal is located and the plane where the top of the train door is located as the lower and upper three-dimensional area when the train is on the platform. Foreign objects refer to human body parts or objects added to the risk space after the platform doors and train doors are closed and ...

AI Technical Summary

Problems solved by technology

[0006] Invention patent application CN106864463A (published on June 20, 2017), using a four-layer laser scanning light curtain installed under the concrete top beam between the platform door and the train (that is, four-layer laser scanning sensors, the angle between layers 2°) for foreign object detection, which can effectively detect foreign objects falling on the scanning surface with four layers of angles of 0°, 2°, 4° and 6°, and can accurately obtain the shape, size and position of obstacles falling on the scanning surface Information, because the laser is too transparent, it can effectively avoid the interference of external light, but because there are large areas of glass in the platform door and train door, there are also shortcomings in the misjudgment of the target caused by the laser passing through the glass and the general defects of laser measurement, such as Due to the extremely low refractive index and reflectivity (such as black and other light-absorbing objects) of the detected objects (such as train doors, platform doors, and stainless steel pedals with relatively smooth surfaces), objects are missed

However, due to the large amount of glass in the train doors and platform doors, the number and arrangement of people waiting in the train compartment and the platform have a great impact on the risk space image. The foreign object recognition system based on visible light is easily affected by ambient light and shadows, especially the sliding doors of the platform. The image of the glass of the train door cannot measure the three-dimensional size of foreign objects

[0013] 2) Single-layer and four-layer laser scanning light curtains are less affected by external light, and can measure the size of objects falling on the scanning plane, but the visibility is poor. Due to the large detection blind zone, they can only detect people and large foreign objects in the risk space. At the same time, there is also the disadvantage that the laser will pass through the glass and cause target misjudgment and the general defect of laser measurement

[0014] 3) The method based on the combination of visible light camera and laser scanning light curtain has greatly improved the false alarm rate and improved accuracy in the detection of people and large foreign objects (which must be able to fall on the scanning plane), but it has no effect on other situations. what effect

[0015] 4) The existing foreign object detection technology or device between the platform door and the train only involves the detection function of foreign objects, the function is single, and it is impossible to accurately measure the three-dimensional size of foreign objects and identify objects. To evaluate whether it will affect driving and safety, the platform door system also needs to provide The platform sliding door switch signal, the train system provides the train door switch signal, and the signal system provides the signal of whether the train is at the station

[0016] 5) The acquisition of RGB images is greatly affected by the external ambient light. When the external light intensity is insufficient at night or in rainy or foggy days, it may cause problems in RGB image acquisition, which is not suitable for the working environment of ground and elevated platforms.

[0017] 6) Compared with the whole image, it is easy to overwhelm the detection of small foreign objects

[0020] Depth image sensors also have multi-layer laser scanning sensors based on mechanical-optical rotating structures. At present, there are 4 layers or 8 layers or 16 layers or 32 layers or 64 layers or higher, which are widely used in long-distance measurement of moving targets, target recognition and three-dimensional Target reconstruction, widely used in obstacle detection in automotive autonomous driving, but the price is high and the reliability is relatively low

Images