Protection method of rail transit station platform door laser and visual technology

Abstract

The invention discloses a protection method of a rail transit station platform door laser and vision technology. The protection method comprises the following steps that a semiconductor laser emits laser beams in a continuous output mode after being collimated and shaped by a beam expander; the laser beam returns scattered light through the environment medium, and the scattered light inhibits the environment light through the band-pass optical filter to form a light spot; the light spots are received by the industrial camera, and after exposure, gain and gamma primary processing are completed, data are converted and transmitted to a PC end through a high-speed link. According to the laser correlation technology, based on the linear propagation characteristic of laser beams, the laser beams are transmitted through the transmitting end, the laser beams are received through the receiving end, and light spots are formed. Meanwhile, a detection method based on vision is adopted, human eyes are replaced by a camera and a computer, a gap between a station door and a train door is monitored in real time, when the foreign matter blocks a characteristic light band, the system judges the existence of the foreign matter through an image processing algorithm, and the device is high in anti-interference capacity, good in stability and high in response speed.

Description

Technical Field

[0001] The present invention relates to the technical field of rail transit safety facilities, and in particular to a protection method for rail transit station platform doors using laser and visual technology. Background Art

[0002] In rail transit systems, securing the gap between platform doors and train doors is crucial for passenger safety. Traditional platform door foreign object detection devices typically utilize infrared beam sensors, which transmit and receive infrared beams to form a detection grid. When a foreign object obstructs the beam, an alarm is triggered. These devices are low-cost and easy to install, making them suitable for simple scenarios. However, they are susceptible to false alarms due to ambient light interference (such as excessive light and dust) and dust obstruction. They also have poor anti-interference capabilities and low detection accuracy.

[0003] This case was created to solve the above problems. Summary of the Invention

[0004] In response to the shortcomings of the existing technology, the present invention provides a protection method for the laser and visual technology of the platform door of a rail transit station, which solves the problems raised in the above background technology. To achieve the above purpose, the present invention is implemented through the following technical solutions: A protection method for the laser and visual technology of the platform door of a rail transit station, comprising the following steps: S1, working end: the semiconductor laser emits a laser beam in a continuous output mode after being collimated and shaped by a beam expander; the laser beam returns scattered light through the ambient medium, and the scattered light is suppressed by a bandpass filter to form a light spot; S2, receiving end: the light spot is received by the industrial camera, and after completing the initial exposure, gain, and gamma processing, the data is converted and transmitted to the PC end through a high-speed link; S3, PC end processing: at startup, the calibration board is first used to complete the distortion correction and brightness calibration; the acquisition thread is separated from the UI thread, and zero-copy feedback is achieved through a ring buffer. When an abnormality occurs, the watchdog is triggered to re-sampling to ensure continuity; the above-mentioned industrial camera and semiconductor laser are located on opposite sides of the platform door (i.e., on both sides of the platform door).

[0005] The laser beam emitted by the semiconductor laser is 940nm. A filter is installed in front of the industrial camera to filter out all light outside the 940nm ± 10nm range. The industrial camera's lens is an infrared camera that collects light in the 700-1100nm range. The opposing laser beam enters the dark chamber, forming a light spot and preventing ambient light from entering. Furthermore, the industrial camera is tilted in the dark chamber, with the lens tilted downward at a 45° angle, effectively reducing the impact of light.

[0006] The above-mentioned industrial cameras and semiconductor lasers are located on opposite sides of the platform door. When a foreign object appears inside the platform door and blocks the characteristic light band (abnormality), the PC determines the presence of the foreign object through image processing and algorithms.

[0007] As a preferred solution, further, the PC-side image processing is as follows: by setting a brightness threshold, the light band area is separated from the background, and then morphological operations are used to remove noise and fill the light band area; when foreign matter blocks the light band, the image after threshold segmentation will have a broken or shadow area, and the presence of foreign matter can be determined by detecting these anomalies.

[0008] As a preferred solution, further, the image processing algorithm is as follows: Spot positioning (sub-pixel) (1) ; (1) Spot centroid: where χ and y represent the horizontal and vertical coordinates of a single pixel in the ROI (in pixels), and I(χ,y) is the grayscale or intensity value of the pixel (0-255 for 8-bit and 0-4095 for 12-bit); The symbol represents the sum of all pixels in the ROI; c ,y c are the horizontal and vertical coordinates of the center of the light spot calculated by the grayscale centroid method, which can achieve sub-pixel accuracy; Spot size estimation (2) ; (2) r is the equivalent radius of the light spot, which is used to measure the size of the light spot; χ c ,y c The center of mass result is taken from formula (1); the other symbols χ, y, I(χ, y) and the summation symbol have the same meaning as formula (1); Morphological Fitting (3) ; (3) Two-dimensional Gaussian fitting: where I(χ,y) is the pixel intensity after filtering or normalization; I0 is the Gaussian peak amplitude (brightness at the center of the spot); (χ0,y0) is the coordinate of the center of the fitted spot; σ is the standard deviation of the Gaussian distribution, which reflects the spot width; Β is the uniform background term, which compensates for ambient light or sensor dark current; Occlusion judgment (4) ; (4) Occlusion determination and debounce: middle, is the adaptive threshold of the t-th frame; and are the average grayscale and standard deviation of the ROI in the frame; k is an empirical coefficient (usually 0.5-1.0). The subsequent exponential sliding average formula middle, is the smoothing factor (0<α≤1), is the EMA value of the previous frame; if N consecutive frames meet St <T t That is to say, the laser is blocked. If the interval between two state flips is less than the minimum jitter time threshold 50 ms) is ignored; Noise Suppression(5): ; (5) Noise suppression: Temporal median filtering removes noise using a 5-frame window; followed by spatial Gaussian convolution middle, is the pixel after time domain filtering, is the convolution kernel displacement index, The standard deviation is A discrete Gaussian kernel is used to smooth the spatial noise.

[0009] The overall concept of the invention: In rail transit systems, the gap between platform doors and train doors is a key factor in ensuring passenger safety. Traditional mechanical or photoelectric sensors have problems such as blind spots and high false alarm rates. The combination of laser protection and visual technology can significantly improve detection accuracy and reliability. A set of laser equipment (transmitter and receiver) is used for mutual emission, and a high-definition industrial camera is deployed to collect light spots. At the same time, the visual algorithm is used to analyze the spot image in real time and identify obstacles.

[0010] By adopting the above technical solution, the present invention provides a laser and vision technology protection method for rail transit station platform doors, which has the following advantages compared to existing technologies: Laser beam transmission technology uses the linear propagation characteristics of laser beams, with the laser beam emitted by the transmitter and received by the receiver, forming a light spot. It also uses a vision-based detection method, using cameras and computers to replace the human eye, to monitor the gap between the platform door and the train door in real time. When foreign objects obstruct the characteristic light band, the system uses image processing algorithms to determine the presence of foreign objects. The device has strong anti-interference capabilities, good stability, and fast response speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 : A flow chart of a protection method for rail transit station platform doors using laser and vision technology. DETAILED DESCRIPTION

[0012] In order to make the purpose, technical solutions and advantages of the present invention more clear, the following is combined with specific embodiments and with reference to the attached drawings. Figure 1 , the present invention is described in further detail.

[0013] Equipment Installation: Align the transmitter and receiver. Adjust the beam to ensure it is unobstructed. Flexible Installation: The laser beam's emission angle and distance can be adjusted as needed to suit different scenarios.

[0014] High precision: The linear propagation characteristics of the laser beam determine its high positioning accuracy, which is suitable for scenarios that require precise detection; strong anti-interference ability: The laser beam is not easily affected by ambient light and electromagnetic interference and has good stability; fast response speed: The laser beam system has a short response time and can quickly detect occlusion events; strong adaptability: It can adapt to complex environments such as lighting changes and background interference; rich information: It can obtain detailed information such as the target's shape, color, texture, etc.

[0015] During the test, the brightness of the collected light spot was compared with a set value (adjustable). Specifically, two beams of light and two light panels were used simultaneously, each with a threshold set. Meeting the set value indicated satisfaction. (In actual operation, the aperture darkened when obstructed, and brightened when unobstructed.) It should be noted that the two beams of light were emitted at different heights, with one high and one low corresponding to tall adults and shorter children, respectively.

[0016] Working Principle: Laser beam detection technology utilizes the rectilinear propagation characteristics of laser beams. Laser beams are emitted from a transmitter and received by a receiver, forming a light spot. Vision-based detection uses cameras and computers to replace the human eye, monitoring the gap between the platform door and the train door in real time. When a foreign object obstructs the characteristic light band, the system uses image processing algorithms to detect its presence.

[0017] This solution's visual algorithm (image processing algorithm) separates the light band from the background by setting a brightness threshold. Morphological operations are then used to remove noise and fill the light band. When a foreign object obstructs the light band, the thresholded image will show breaks or shadows. Detecting these anomalies can identify the presence of a foreign object. Deep learning and video analysis can also be combined to improve foreign object detection accuracy.

[0018] The light spot is collimated and shaped by a beam expander from a semiconductor laser before being emitted in a continuous output mode. Scattered light returning from the working medium is first filtered by a bandpass filter to suppress ambient light before being received by an industrial camera (CMOS, global shutter) through a C-mount lens. The camera's FPGA performs preliminary processing, including exposure, gain, and gamma, before converting the RAW data into 8 / 12-bit RGB and transmitting it to a PC via a high-speed USB 3.0 link. The host computer SDK maps the frame buffer to the acquisition software, which then uses DMA data streaming for real-time display, synchronous archiving, and subsequent algorithm processing.

[0019] At startup, a calibration board is used to perform distortion correction and brightness calibration. The acquisition thread is separated from the UI thread, achieving zero-copy through a ring buffer. A watchdog resampling function triggers abnormalities to ensure continuity. The entire process has a latency of less than 30ms, meeting dynamic monitoring requirements.

[0020] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above are only specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A laser and visual technology protection method for platform doors of rail transit stations, characterized in that: The steps include: S1, PC processing: When starting, first use the calibration board to complete distortion correction and brightness calibration; acquisition thread Separated from the UI thread, zero-copy feedback is achieved through a ring buffer; S2, working end: the semiconductor laser is collimated and shaped by the beam expander and then emits laser light in a continuous output mode. Light beam; The laser beam returns scattered light through the ambient medium, and the scattered light passes through a bandpass filter to suppress the ambient light to form a light spot; S3, receiving end: the light spot is received by the industrial camera and after the initial exposure, gain and gamma processing is completed, After data conversion, it is transmitted to the PC via a high-speed link; S4, when a foreign object blocks the characteristic light band inside the platform door - when abnormal, the PC side processes the image The algorithm determines the presence of foreign objects and triggers the watchdog to resample to ensure continuity; The above-mentioned industrial camera and semiconductor laser are located on the opposite side of the platform door.

2. The laser and visual technology protection method for rail transit station platform doors according to claim 1, characterized in that: The laser beam emitted by the semiconductor laser is 940nm, and a filter is installed in front of the industrial camera to filter out light outside 940nm±10nm.

3. The laser and visual technology protection method for platform doors of rail transit stations according to claim 2 is characterized in that :The camera of the industrial camera is an infrared camera that collects light in the range of 700-1100nm.

4. The laser and visual technology protection method for rail transit station platform doors according to claim 3 is characterized in that The lens side of the industrial camera is tilted downward and placed in a dark chamber with an inclination angle of 45 degrees. The laser beam on the opposite side is shot into the dark chamber to form a light spot.

5. The laser and visual technology protection for rail transit station platform doors according to claim 1 The method is characterized in that: The image processing on the PC side is as follows: by setting the brightness threshold, the light band area is separated from the background, and then morphological operations are used to remove noise and fill the light band area; When foreign objects block the light band, the image after threshold segmentation will have broken or shadow areas. By detecting these anomalies, the presence of foreign objects can be determined.

6. The laser and visual technology protection for rail transit station platform doors according to claim 1 The method is characterized in that The image processing algorithm is as follows: Spot positioning (sub-pixel) (1): ； (1) Spot centroid: where χ and y represent the horizontal and vertical coordinates of a single pixel in the ROI (in pixels), and I(χ,y) is the grayscale or intensity value of the pixel (0-255 for 8-bit and 0-4095 for 12-bit); The symbol represents the sum of all pixels in the ROI; c ,y c are the horizontal and vertical coordinates of the center of the light spot calculated by the grayscale centroid method, which can achieve sub-pixel accuracy; Spot size estimation (2): ； (2) r is the equivalent radius of the light spot, which is used to measure the size of the light spot; χ c ,y c The center of mass result is taken from formula (1); the other symbols χ, y, I(χ, y) and the summation symbol have the same meaning as formula (1); Morphological fitting (3): ； (3) Two-dimensional Gaussian fitting: where I(χ,y) is the pixel intensity after filtering or normalization; I0 is the Gaussian peak amplitude (brightness at the center of the spot); (χ0,y0) is the coordinate of the center of the fitted spot; σ is the standard deviation of the Gaussian distribution, which reflects the spot width; Β is the uniform background term, which compensates for ambient light or sensor dark current; Occlusion judgment (4): ； (4) Occlusion determination and debounce: middle, is the adaptive threshold of the t-th frame; and are the average grayscale and standard deviation of the ROI in the frame; k is an empirical coefficient (usually 0.5-1.0). The subsequent exponential sliding average formula middle, is the smoothing factor (0<α≤1), is the EMA value of the previous frame; If N consecutive frames satisfy S t <T t That is to say, the laser is blocked. If the interval between two state flips is less than the minimum jitter time threshold 50 ms) is ignored; Noise Suppression(5): ； (5) Noise suppression: Temporal median filtering removes noise using a 5-frame window; followed by spatial Gaussian convolution middle, is the pixel after time domain filtering, is the convolution kernel displacement index, The standard deviation is A discrete Gaussian kernel is used to smooth the spatial noise.

Images