Accurate camera triggering method based on vehicle speed prediction and image acquisition system

Abstract

The invention discloses a vehicle speed prediction-based camera accurate triggering method and an image acquisition system, and the system comprises a speed measurement subsystem which is used for measuring the speed of a train; comprising a high-speed detection sensor group and a low-speed detection sensor group; the high-speed detection sensor group is used for detecting the speed of a train with the speed higher than a speed threshold value, and the low-speed detection sensor group is used for detecting the speed of a train with the speed smaller than or equal to the speed threshold value; the controller is used for predicting the arrival time of the train arriving at a preset acquisition position according to the train speed acquired by the speed measurement subsystem, and calculating the image acquisition triggering time of the image acquisition module according to the arrival time; issuing an instruction for carrying out image acquisition at the image acquisition triggering time to the shooting module; the system further comprises an image acquisition module. According to the invention, accurate speed measurement can be carried out on trains with different speeds so as to predict the time when the trains arrive at the preset shooting place, so that the cameras are accurately triggered to collect train wheel images.

Description

Technical Field

[0001] The present invention belongs to the technical field of train detection, and particularly relates to a precise camera triggering method based on vehicle speed prediction and an image acquisition system. Background Art

[0002] Intelligent operation and maintenance has become an important direction for the development of railways. Intelligent detection is a key link in realizing railway intelligent operation and maintenance, and can provide tool means and data support for the safe operation, efficient operation, and rapid maintenance of trains. As a key component to ensure the safe operation of trains, for a long time, the detection of key components of train car bodies in China's railway departments has mostly adopted manual interpretation or parking detection, which has problems such as low efficiency, high cost, and safety hazards.

[0003] The prior art Chinese patent application CN202111672028.6 discloses a train vehicle speed acquisition and camera triggering method based on multi-source data fusion, including the following steps: Step 1: Establish a wheel speed stereo vision measurement model; Step 2: Establish a vehicle speed measurement model based on a magnetic steel array; Step 3: Fusion of speed measurement and triggering based on the Kalman filter method; The measured speed is filtered by the Kalman filter method to eliminate gross errors, and a stable final speed value is obtained after weighted fusion: After the magnetic steel detects the train, the train command is started. The wheels pass by the magnetic steel in sequence. After each magnetic steel detects the train data, it is transmitted to the PLC. After the image speed measurement subsystem detects the train data, it is transmitted to the PLC. The PLC receives the train information including the magnetic steel speed measurement value and the image speed measurement value and performs analysis and processing. After weighted fusion, the final speed value is obtained, and a high-frequency pulse is generated to trigger the binocular camera to collect train images.

[0004] In the above prior art, after obtaining the train speed by establishing a vision measurement model and a magnetic steel array speed measurement model respectively and then fusing them, the accurate train speed is obtained to accurately trigger the camera to collect train images. This prior art requires establishing an additional vision model, which has a high cost and is difficult to implement. Summary of the Invention

[0005] The purpose of the present invention is to provide a precise camera triggering method based on vehicle speed prediction and an image acquisition system, which partially solves or alleviates the above deficiencies in the prior art, and can accurately measure the speed of trains at different speeds so as to predict the time when the train reaches the preset shooting location, thereby accurately triggering the camera to collect train wheel images.

[0006] To solve the above-mentioned technical problems, the present invention specifically adopts the following technical solutions: A train image acquisition system, including: A speed measurement subsystem for measuring the speed of a train, including a high-speed detection sensor group and a low-speed detection sensor group. The high-speed detection sensor group is used to detect the speed of trains with speeds higher than the speed threshold, and the low-speed detection sensor group is used to detect the speed of trains with speeds less than or equal to the speed threshold. A controller for predicting the arrival time of the train at a preset acquisition position using the train speed obtained by the speed measurement subsystem, and calculating the image acquisition trigger time of the image acquisition module according to the arrival time; issuing an instruction to the shooting module to perform image acquisition at the image acquisition trigger time. An image acquisition module for performing image acquisition in response to the instruction issued by the controller.

[0007] As an improvement, the high-speed detection sensor group includes wheel sensors arranged before and after along the track; and a data acquisition box, which calculates the speed of the train according to the distance between the front and rear wheel sensors and the time difference between the front and rear wheel sensors detecting the same wheel.

[0008] As an improvement, the low-speed detection sensor group includes a sensor array composed of at least two photoelectric sensors; the photoelectric sensors include a transmitting end and a receiving end respectively arranged on both sides of the track; the light rays emitted by several transmitting ends in the sensor array form a light curtain for detecting the traveling speed of the wheels.

[0009] As an improvement, the low-speed detection sensor group detects the train and judges the traveling speed range of the train.

[0010] The present invention also provides a method for accurately triggering a camera based on vehicle speed prediction, which is applied to the above-mentioned train image acquisition system, including: Judging whether a train is coming, and judging the direction and speed range of the train; If the train speed is greater than the set speed threshold, the high-speed detection sensor group is selected to detect the speed of the train, otherwise the low-speed detection sensor group is selected to detect the speed of the train; Predicting the time when the train arrives at the preset acquisition position using the detected train speed; Triggering the image acquisition module to acquire the train wheel image according to the arrival time.

[0011] As an improvement, the method for the high-speed detection sensor group to detect the speed of the train includes: Judging whether the wheel sensor in front of the train direction is triggered; if triggered, judging the validity of the trigger signal; When the trigger signal is valid, judging whether the speed measurement process times out; if not timed out, judging whether the wheel sensor behind the train direction is triggered; if triggered, judging the validity of the trigger signal; When the trigger signal is valid, calculate the current speed of the train based on the distance between two wheel sensors and the time difference between the front and rear wheel sensors being triggered by the same wheel.

[0012] As an improvement, the method for the low-speed detection sensor group to detect the speed of the train includes: Judging in sequence from the train direction to the outgoing train direction whether the photoelectric sensors are triggered. If triggered, judge the validity of the trigger signal; When the trigger signal is valid, calculate the current speed of the train based on the distance between any two photoelectric sensors and the time difference between the two photoelectric sensors being triggered by the same wheel.

[0013] As an improvement, after calculating the current speed of the train, judge whether the speed meets the train speed trend; If not, correct the current speed.

[0014] As an improvement, the method for predicting the time when the train arrives at a preset acquisition position includes: When the current vehicle speed is higher than the speed threshold, obtain the delay of the wheel sensor according to the current train speed; Use the formula t1 = s1 / v to calculate the theoretical time for the train to reach the preset acquisition position, where t1 is the theoretical time, s1 is the distance between the wheel sensor behind the train direction and the preset acquisition position, and v is the current speed of the train; Use the formula t = t1 - t2 to calculate the time for the train to reach the preset acquisition position, where t is the arrival time, t1 is the theoretical time, and t2 is the delay time of the wheel sensor; When the current vehicle speed is lower than or equal to the speed threshold, obtain the delay of the non-photoelectric sensor according to the current train speed; Use the formula t1 = s2 / v to calculate the theoretical time for the train to reach the preset acquisition position, where t1 is the theoretical time, s2 is the distance between the last triggered photoelectric sensor and the preset acquisition position, and v is the current speed of the train; Use the formula t = t1 - t3 to calculate the time for the train to reach the preset acquisition position, where t is the arrival time, t1 is the theoretical time, and t3 is the delay time of the photoelectric sensor.

[0015] As an improvement, check whether there is a fault in the photoelectric sensor; When a certain photoelectric sensor has a fault, shield the photoelectric sensor and then reorder all the photoelectric sensors; Reset the distance between adjacent photoelectric sensors.

[0016] Beneficial effects: According to the different characteristics of the sensors, different speed detection sensor groups are selected to detect the train speed in different speed ranges, so as to meet the detection requirements. Its speed adaptation range is wide (0 - 120 km / h), and it can adapt to situations such as two-way, high-speed, low-speed, parking, reversing, and creeping.

[0017] When used as the triggering mechanism for image acquisition, the image acquisition trigger position is accurate, and the position consistency of the components in the image is relatively good. By calculating the sensor delay based on the speed, the time when the train arrives at the preset acquisition position can be accurately predicted, ensuring the accuracy of the image acquisition timing and improving the image acquisition quality.

[0018] In addition, the optoelectronic sensors are equipped with a fault replacement mechanism, which can timely shield the faulty sensors, reorder them, and set the adjacent distances, ensuring the accuracy of speed detection. At the same time, the sensor signals have self-diagnosis and substitution logic, which can locate the faulty signal points, enable adjacent signal substitution and data interpolation, and ensure the stable operation of the system.

[0019] From speed measurement, signal processing to image acquisition, each link of the present invention is closely related. By accurately controlling the operation of the camera and light source according to the train speed and signal status, the system can achieve efficient collaborative operation, enhancing the reliability and practicality of the entire train image acquisition system. BRIEF DESCRIPTION OF THE DRAWINGS In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to actual scale. Obviously, the following-described drawings are some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0020] Figure 1 It is the structural schematic diagram of the present invention.

[0021] Figure 2 It is the flow chart of the present invention.

[0022] Figure 3 It is the flow chart of speed measurement and predicting the arrival time in the present invention.

[0023] Figure 4 It is the flow chart of judging the signal validity in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0025] In this document, suffixes such as "module", "component" or "unit" used to represent elements are only for the convenience of describing the present invention, and have no specific meaning in themselves. Therefore, "module", "component" or "unit" can be used interchangeably.

[0026] In this document, terms such as "upper", "lower", "inner", "outer", "front", "rear", "one end", "the other end", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be construed as a limitation of the present invention. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0027] In this document, unless otherwise clearly defined and limited, terms such as "installation", "provided with", "connection", etc. should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, a direct connection, or an indirect connection through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0028] In this document, "and / or" includes any and all combinations of one or more of the listed related items.

[0029] In this document, "a plurality of" means two or more, that is, it includes two, three, four, five, etc.

[0030] Embodiment 1: As Figure 1 shown, the present invention provides a train image acquisition system, including: A speed measurement subsystem for measuring the speed of the train; including a high-speed detection sensor group and a low-speed detection sensor group; the high-speed detection sensor group is used for speed detection of trains with a speed higher than the speed threshold, and the low-speed detection sensor group is used for speed detection of trains with a speed less than or equal to the speed threshold; A controller is configured to predict the arrival time of a train at a preset acquisition position by using the train speed obtained by a speed measurement subsystem, and calculate the image acquisition trigger time of an image acquisition module according to the arrival time; and send an instruction to a shooting module to perform image acquisition at the image acquisition trigger time. An image acquisition module is configured to perform image acquisition in response to an instruction sent by the controller. In this embodiment, the image acquisition module includes a total of 8 cameras C1 to C8. Among them, C1, C3, C5, and C7 form a group for acquiring the outer and inner images of one wheel. C2, C4, C6, and C8 form a group for acquiring the outer and inner images of the other wheel on the same axis. Of course, the number and arrangement of the cameras can be adjusted according to specific situations and are not limited in the present invention.

[0031] Due to the characteristics of sensors, existing speed sensors cannot achieve effective coverage in the full speed range. Therefore, in the speed measurement subsystem of the present invention, according to the different characteristics of sensors, different speed detection sensor groups are selected to detect the train speed in different speed intervals, so as to meet the detection requirements.

[0032] Specifically, in the present invention, a high-speed detection sensor group and a low-speed detection sensor group are provided. When the speed of the train is greater than or equal to a speed threshold, such as 3 km / h, the high-speed detection sensor group is selected for detection. When the speed of the train is lower than the speed threshold, the low-speed detection sensor group is selected for detection.

[0033] In addition, the low-speed detection sensor group is also used to detect an oncoming train and judge the driving speed range of the oncoming train; the controller can also select the high-speed detection sensor group and the low-speed detection sensor group to detect the driving speed of the oncoming train according to the driving speed range of the oncoming train.

[0034] More specifically, in the present invention, the high-speed detection sensor group includes wheel sensors (the wheel sensors can be magnetic steel) arranged along the track before and after; and a data acquisition box, and the data acquisition box calculates the speed of the train according to the distance between the front and rear two wheel sensors and the time difference between the front and rear two wheel sensors detecting the same wheel.

[0035] The wheel sensors can sense the wheels when the wheels pass through the detection area, and thus send signals indicating that the wheels are detected to the data acquisition box. According to the time difference between the signals of the front and rear two wheel sensors detecting the same wheel and the distance between the two wheel sensors, the speed of the train can be calculated.

[0036] In addition, in order to facilitate the detection of oncoming trains in two directions, in the present invention, the high-speed detection sensor group is two groups respectively arranged on two tracks; as Figure 1As shown, wheel sensors WS01 and WS02 form a group (connected to data acquisition box 2) for detecting oncoming trains from direction B. Wheel sensors WS03 and WS04 form a group (connected to data acquisition box 3) for detecting oncoming trains from direction A. The wheel sensors can be arranged inside the track or outside the track, and can be flexibly adjusted according to the actual situation.

[0037] The two groups of high-speed detection sensor groups respectively detect the vehicle speeds of trains coming from two directions. For the two groups of high-speed detection sensor groups, the wheel sensors at the rear are within the detection area of the low-speed detection sensor group, so that the detection areas of the high-speed detection sensor group and the low-speed detection sensor group partially overlap. That is to say, no matter which direction the oncoming train is from, the train will first trigger one of the wheel sensors at the front of one of the groups of high-speed detection sensor groups, and then enter the detection area of the low-speed detection sensor group.

[0038] The reason for such an arrangement is to improve the accuracy of high-speed detection. The distance between the front and rear wheel sensors can be greater than the diameter of the wheel, so as to leave enough spacing to improve the detection accuracy. In order not to let the entire detection area cover too wide a range, the detection areas of the high-speed detection sensor group and the low-speed detection sensor group are made to overlap partially.

[0039] Although the wheel sensors have high accuracy when detecting high-speed trains, when the vehicle speed is too low, the train is stopped or reversed, the signals will be weak or unstable, making the position judgment inaccurate. Therefore, in the present invention, a sensor array composed of at least two photoelectric sensors is selected as the low-speed detection sensor group for low-speed detection of trains. The photoelectric sensors include a transmitting end and a receiving end respectively arranged on both sides of the track; the light rays emitted by several transmitting ends in the sensor array form a light curtain for detecting the traveling speed of the wheel.

[0040] After the train travels into the range of the light curtain, it can block a part of the light rays in the light curtain, causing the output electrical signal to change, thereby detecting the vehicle speed. Since the entire light curtain is used to detect the wheel, whether the train stops, reverses, or creeps within the detection area, it will be accurately detected.

[0041] Of course, the detection method using the light curtain is prone to interference by other tangible foreign objects and has a relatively high internal delay, so it is only suitable for low-speed detection.

[0042] More specifically, the connection line between the transmitting end (connected to data acquisition box 4) and the receiving end (connected to data acquisition box 1) is perpendicular to the track, that is, the light rays are perpendicular to the track axis, which is more convenient for detecting the wheel. In addition, several photoelectric sensors are evenly arranged in the front-rear direction of the track, and the width of the formed light curtain is greater than the diameter of the wheel. Of course, existing light curtain sensors (whose principle is the same as that of the photoelectric sensor array) can also be selected.

[0043] The present invention realizes precise speed measurement of a train within the full speed range by using a high-speed detection sensor group and a low-speed detection sensor group, effectively avoiding the drawbacks of the two detection sensor groups. On the premise of precise speed measurement, the time when the train arrives at the preset acquisition area is precisely predicted and the image acquisition module is triggered, so that trains at any speed can collect images with high consistency.

[0044] Embodiment 2: As Figure 2 shown, the present invention also provides a precise camera triggering method based on vehicle speed prediction, which is applied to the above train image acquisition system and includes: S1 Determine whether a train is coming, and determine the direction and speed range of the train.

[0045] Specifically, this step is executed by using a low-speed detection sensor group. The low-speed detection sensor group consists of a sensor array composed of at least two photoelectric sensors, and its transmitting end and receiving end are respectively arranged on both sides of the track. The light rays emitted by several transmitting ends form a light curtain. When the train travels within the range of the light curtain, it will block some of the light rays in the light curtain, causing the output electrical signal to change. The system determines that a train is coming by detecting this change in the electrical signal.

[0046] Since the photoelectric sensors are evenly arranged in the front-back direction of the track, the direction of the train can be determined from the position of the first triggered sensor when the train enters the light curtain, combined with the arrangement order of each sensor. For example, if the photoelectric sensor in front of the track is triggered first, it can be determined that the train is coming in that direction.

[0047] Judge whether the photoelectric sensors are triggered in sequence from the train direction to the departure direction of the train. When the trigger signal is valid, calculate the current speed of the train according to the distance between any two photoelectric sensors triggered by the same wheel and the trigger time difference. Compare the calculated speed with the set speed threshold (such as 3 km / h) to determine the speed range of the train, providing a basis for selecting a suitable speed measurement sensor group in the subsequent process.

[0048] The reason for choosing the low-speed detection sensor group to execute this step is as follows: First of all, the light curtain detection method formed by the photoelectric sensor array detects the wheels through the entire light curtain, and can provide a stable and reliable detection signal when the train is running at a low speed. Compared with the unstable performance of the wheel sensor at low speed, it is more suitable for detecting the low-speed state at the initial stage when the train enters the detection area.

[0049] Secondly, the optoelectronic sensors are evenly arranged in the front-back direction of the track. From the position of the first triggered sensor when the train enters the light curtain and in combination with the arrangement order of each sensor, the running direction of the train can be conveniently judged. This layout and detection method have natural advantages for direction judgment.

[0050] S2 If the train speed is greater than the set speed threshold, the high-speed detection sensor group is selected to detect the train speed; otherwise, the low-speed detection sensor group is selected to detect the train speed.

[0051] Specifically, in the present invention, a high-speed detection sensor group and a low-speed detection sensor group are provided. When the train speed is greater than or equal to the speed threshold, for example, 3 km / h, the high-speed detection sensor group is selected for detection. When the train speed is lower than the speed threshold, the low-speed detection sensor group is selected for detection.

[0052] As Figure 3 shown, if the high-speed detection sensor group is selected, the speed is measured according to the following steps: S201 Judge whether the wheel sensor in front of the train direction is triggered. If triggered, judge the validity of the trigger signal.

[0053] The wheel sensors in the high-speed detection sensor group are arranged beside the track. When the train wheel approaches and passes the wheel sensor in front of the train running direction, the sensor will sense the presence of the wheel and thus generate a trigger signal.

[0054] As Figure 1 shown by the oncoming train in A, the so-called wheel sensor in front of the train direction is WS01.

[0055] Due to possible interference factors in the actual environment, such as electromagnetic interference and false triggering of nearby objects, it is necessary to judge the validity of the trigger signal. Through a specific algorithm or preset rules, check the characteristics of the trigger signal, such as whether the amplitude, waveform, duration, etc. of the signal conform to the signal pattern of normal wheel triggering. If the signal does not conform to the expected pattern, the trigger signal is considered invalid and no subsequent speed measurement calculation is performed to avoid inaccurate speed measurement results caused by incorrect signals.

[0056] Figure 4 shows the specific process of signal validity judgment.

[0057] First, judge whether the received signal is a valid analog signal. Valid analog signal processing path: If it is determined to be a valid analog signal (Y), the signal will successively pass through a second-order low-pass filter, the purpose of which is to filter out high-frequency noise and allow low-frequency signals to pass through smoothly, thereby making the signal purer. Then, the signal is amplified to enhance its intensity for subsequent processing. Finally, it is output through signal hysteresis comparison, which can effectively avoid unstable output when the signal fluctuates near the threshold.

[0058] After hysteresis comparison, it enters the signal sampling step to digitally collect the signal. Subsequently, signal dynamic filtering and noise reduction are performed to further remove interference components in the signal and improve signal quality. Then, the signal level state time is recorded, which helps analyze the level changes of the signal at different times. Then, the configuration parameters of the signal level state and speed are updated, and the speed and level state information are stored to provide data support for subsequent signal judgment and system control.

[0059] While recording the signal level state time, the existing speed and level state parameters are called to determine again whether it is a valid level signal. If it is not a valid level signal (N), it will return to the judgment process of whether it is a valid analog signal and continue to be processed; if it is a valid level signal (Y), the wheel position is predicted and a valid signal is output, and finally the camera and light source are controlled to start working to achieve the actual application purpose of signal processing.

[0060] S202 If the trigger signal is valid, determine whether the speed measurement process times out.

[0061] To ensure the timeliness and effectiveness of the speed measurement process, the system sets a speed measurement timeout mechanism. During the normal running of the train, the time interval from the triggering of the previous wheel sensor to the triggering of the next one should be within a reasonable range. If this time is too long, it may mean that the train has an abnormal situation (such as sudden stop, sensor failure, etc.), or the signal is lost due to interference, making the speed measurement process unable to proceed normally. Therefore, the process cannot wait indefinitely. In the case of process timeout, the process needs to be restarted.

[0062] More specifically, the system will preset a reasonable time threshold as the standard for timeout judgment. When the trigger signal of the first wheel sensor is determined to be valid, the system starts timing. When the timing exceeds this preset time threshold, it is considered that the speed measurement process times out. If it times out, it means that there may be problems in this speed measurement process and it needs to be restarted or troubleshooting needs to be carried out to ensure the accuracy of the speed measurement result.

[0063] S203 If it does not time out, determine whether the wheel sensor behind the train direction is triggered. If triggered, determine the validity of the trigger signal.

[0064] In the same example as above, as shown in A towards the oncoming vehicle, the so-called wheel sensor that is farther back relative to the direction of the train is WS02.

[0065] When the WS01 trigger signal is confirmed to be valid and the speed measurement process has not timed out, the system waits and detects whether WS02 is triggered. Only when both the front and rear sensors are triggered by the wheels can the complete information required to calculate the train speed be obtained, that is, the time difference between the wheels passing the two sensors.

[0066] Just like the validity judgment of the WS01 trigger signal in the first step, the validity of the WS02 trigger signal also needs to be confirmed. This is because even if the previous signal is valid and has not timed out, the latter signal may still be disturbed and cause errors. Check again whether the characteristics of the trigger signal meet expectations, ensure that the signal truly reflects the situation of the wheel passing the rear sensor, and provide a reliable data basis for accurately calculating the vehicle speed.

[0067] S204 When the trigger signal is valid, the current speed of the train is calculated according to the distance between the two wheel sensors and the time difference between the front and rear wheel sensors being triggered by the same wheel.

[0068] In this step, the current speed of the train is calculated based on the speed calculation formula in physics. In this scenario, the distance between the two wheel sensors is a known fixed value, and the time difference between the front and rear wheel sensors being triggered by the same wheel is obtained through system timing. Substituting these two data into the formula, the current speed of the train between passing the two sensors can be calculated.

[0069] By accurately measuring the time difference between the train passing two fixed position sensors and combining it with the known sensor spacing, the speed of the train in this section can be accurately calculated. This speed information is crucial for the subsequent prediction of the time when the train arrives at the preset acquisition position, and for controlling the image acquisition module to acquire the train wheel image at the right time, ensuring that the entire train image acquisition system can operate according to the predetermined process and accuracy requirements.

[0070] The principle of low-speed detection sensor group to detect train speed is similar to that of high-speed detection sensor group. If low-speed detection sensor group is selected, speed measurement is performed according to the following steps: S211 determines whether the photoelectric sensor is triggered from the train direction to the outgoing direction, and if triggered, determines the validity of the trigger signal.

[0071] The low-speed detection sensor group consists of multiple photoelectric sensors evenly arranged along the front and rear directions of the track, forming a light curtain for detecting trains. When the train enters the light curtain range, the wheels will block the light in turn, thereby triggering the corresponding photoelectric sensors. Starting from the direction of the train, each photoelectric sensor is detected in sequence to see if it is triggered, so that the process information of the train passing through the light curtain can be accurately recorded.Figure 1 Taking the oncoming vehicle in the middle as an example, the wheels will block the light from front to back in sequence, thus triggering the photoelectric sensors.

[0072] It can be foreseen that, similar to the signal judgment of the high-speed detection sensor group, there are many interference factors in the actual environment, which may cause the photoelectric sensors to generate false trigger signals. To ensure the accuracy of speed measurement, it is necessary to judge the validity of the trigger signals. The system will check various characteristics of the trigger signals according to the preset rules or algorithms, such as whether the amplitude, duration, waveform, etc. of the signals conform to the signal pattern of normal triggering. If the signal does not meet the expectation, it is determined that the trigger signal is invalid, so as to avoid affecting the speed measurement result due to the wrong signal.

[0073] When the trigger signal is valid, S212 calculates the current speed of the train according to the distance between any two photoelectric sensors and the time difference between the two photoelectric sensors being triggered by the same wheel.

[0074] This step still calculates the current speed of the train based on the basic formula of speed. In the scenario of the low-speed detection sensor group, it is known that the distance between any two adjacent photoelectric sensors is a fixed value, and the time difference between the train wheels triggering these two photoelectric sensors can be obtained through the timing function of the system. Using these two key data, the current speed of the train passing between these two sensors can be calculated.

[0075] It can be foreseen that two adjacent photoelectric sensors can be selected to calculate the speed of the train, or photoelectric sensors at other positions can be selected to calculate the speed of the train. The principle is the same and is not limited in the present invention. And due to the diversity of sensor selection, even if some sensors fail or the signals are abnormal, as long as there are still two effectively triggered sensors and the distance between them is known, the speed can be calculated. This design improves the reliability of speed measurement of the system in complex situations, ensures that the speed of the train can be accurately measured in the low-speed driving state, and provides a reliable speed basis for subsequent operations such as train image acquisition.

[0076] In some embodiments, after calculating the current speed of the train, it is also necessary to judge whether the speed meets the train speed trend; if not, the current speed is corrected.

[0077] During the normal operation of the train, its speed change is usually continuous and follows a certain pattern. For example, the acceleration or deceleration process is gradual and there will be no sudden large jumps. By judging whether the currently calculated train speed meets the train speed trend, possible abnormal situations can be detected in a timely manner, such as incorrect speed data caused by sensor failures, calculation deviations caused by external interferences, etc. If the current speed does not match the train's speed trend, it needs to be corrected to ensure the accuracy of various subsequent controls and decisions based on speed data (such as determining the image acquisition timing, etc.).

[0078] In this embodiment, correction can be based on adjacent sensor data. Since there are multiple sensors for speed measurement, the data measured by adjacent sensors can be used for mutual verification and correction. For example, if the speed calculated by a certain sensor is significantly different from the speeds calculated by other adjacent sensors and does not meet the speed trend, its correction can be made with reference to the speed data of adjacent sensors. A weighted average method can be adopted, and different weights are assigned according to factors such as the distance between sensors and measurement accuracy to calculate a more reasonable speed value.

[0079] S3 Predict the time when the train arrives at the preset acquisition position using the detected train speed.

[0080] If the current vehicle speed is higher than the speed threshold, that is, in the case of selecting the high-speed detection sensor group, the prediction steps include: S301 Obtain the delay of the wheel sensor according to the current train speed.

[0081] When the wheel sensor detects the passing of the wheel, from the actual arrival of the wheel at the sensor detection area to the generation of a signal by the sensor and its transmission to the system for processing, this process is not instantaneous but has a certain time delay. This delay is affected by various factors such as the characteristics of the sensor itself, the signal transmission line, and the system processing speed.

[0082] Different train speeds may affect the sensor delay. Generally speaking, the change in train speed may cause the time interval between the wheels passing the sensor to change, thus affecting the entire process of sensor detection and signal processing, resulting in different delay times. Therefore, it is necessary to obtain the corresponding wheel sensor delay according to the actual current train speed.

[0083] In this embodiment, the delay of the sensor at different speeds is obtained by looking up a table.

[0084] S302 Use the formula t1 = s1 / v to calculate the theoretical time for the train to reach the preset acquisition position, where t1 is the theoretical time, s1 is the distance between the wheel sensor behind the train direction and the preset acquisition position, and v is the current train speed.

[0085] The purpose of this step is to calculate the theoretical time for the train to reach the preset acquisition position. It can be foreseen that due to the existence of sensor delay, this theoretical time is not accurate and needs to be further corrected.

[0086] S303 uses the formula t = t1 - t2 to calculate the time for the train to reach the preset acquisition position, where t is the arrival time, t1 is the theoretical time, and t2 is the delay time of the wheel sensor.

[0087] Since the theoretical time t1 calculated in the previous step does not consider the delay t2 of the wheel sensor, in actual situations, this delay will cause the actual arrival time of the train at the preset acquisition position to be earlier than the theoretical time. Therefore, it is necessary to subtract the sensor delay time t2 from the theoretical time t1 to obtain a more accurate arrival time t of the train at the preset acquisition position.

[0088] If the current train speed is lower than or equal to the speed threshold, that is, in the case of selecting the low-speed detection sensor group, the prediction steps include: S311 obtains the delay of the non-photoelectric sensor according to the current train speed.

[0089] Similarly, when the photoelectric sensor detects the train, from the moment the train wheel blocks the light and changes the sensor output electrical signal to the moment the system receives and processes this signal, a certain time delay will occur during this process. This delay is related to factors such as the working principle of the photoelectric sensor, the signal transmission line, and the signal processing speed of the system.

[0090] When the train speed is different, the time characteristics of passing through the photoelectric sensor light curtain are different, which may affect the sensor detection and signal processing process, and thus cause the delay time to change. Therefore, it is necessary to determine the corresponding photoelectric sensor delay according to the actual current train speed.

[0091] In this embodiment, the delay of the sensor at different speeds is obtained by looking up a table.

[0092] S312 uses the formula t1 = s2 / v to calculate the theoretical time for the train to reach the preset acquisition position, where t1 is the theoretical time, s2 is the distance between the last triggered photoelectric sensor and the preset acquisition position, and v is the current train speed.

[0093] The purpose of this step is to calculate the theoretical time for the train to reach the preset acquisition position. It can be foreseen that due to the existence of sensor delay, this theoretical time is not accurate and needs to be further corrected.

[0094] S313 uses the formula t = t1 - t3 to calculate the time for the train to reach the preset acquisition position, where t is the arrival time, t1 is the theoretical time, and t3 is the delay time of the photoelectric sensor.

[0095] Since the theoretical time t1 calculated in the previous step does not consider the delay of the optoelectronic sensor. In fact, due to the delay of the optoelectronic sensor, the actual arrival time of the train at the preset acquisition position will be earlier than the theoretical time. Therefore, the delay time t3 of the optoelectronic sensor should be subtracted from the theoretical time t1 to obtain a more realistic arrival time of the train at the preset acquisition position.

[0096] S4 triggers the image acquisition module to acquire the train wheel image according to the arrival time.

[0097] The system accurately triggers the image acquisition module when the train is about to reach the preset acquisition position according to the time t, ensuring that clear images of the train wheels can be acquired at the best time, meeting the system's requirements for the accuracy and timeliness of image acquisition, so that the images acquired at any vehicle speed can be at the same position.

[0098] It should be noted that the trigger delay of the image acquisition module is very small and will not affect the acquisition. Therefore, it is not considered in the present invention.

[0099] In addition, for the low-speed detection sensor group, since multiple optoelectronic sensors can be set. The failure of one or a part of the optoelectronic sensors will not affect the overall speed measurement work. Therefore, an optoelectronic sensor replacement strategy is also provided in this embodiment: First, check whether the optoelectronic sensor is faulty.

[0100] In this embodiment, various methods can be used to check for faults. For example, continuously monitor the stability of the sensor output signal. Under normal circumstances, the optoelectronic sensor outputs a stable signal when the train does not block the light and another stable signal when blocked. If the signal shows abnormal fluctuations, exceeds the normal range, or does not change for a long time, it may indicate that the sensor is faulty. In addition, it is also possible to conduct comparative analysis with the data of other sensors. If it is found that the data of a certain optoelectronic sensor is significantly different from the train operation situation reflected by other sensors, it may also imply that the sensor is faulty. At the same time, the system may also regularly perform a self-check program, actively send a detection signal to the optoelectronic sensor, and check whether its feedback is normal.

[0101] Then, in the case of a faulty optoelectronic sensor, shield the faulty optoelectronic sensor and reorder all the optoelectronic sensors.

[0102] Once it is determined that a certain optoelectronic sensor has a fault, in order to avoid interference caused by its incorrect data to the system, it is necessary to shield it from the sequence of sensors operating normally. This means that the system will no longer incorporate the data of this faulty sensor into subsequent processing procedures such as speed calculation and train position determination. The shielding operation can be achieved through the switch control of the hardware circuit or the setting of the sensor data acquisition channel by software.

[0103] After shielding the faulty sensor, in order to ensure that the system can accurately judge the running state of the train according to the trigger order of the remaining normal sensors, it is necessary to reorder all the remaining optoelectronic sensors. This is because the order has changed after removing one sensor from the sensors originally arranged in a certain order. Reordering allows the system to continue processing the sensor trigger signals according to a coherent logic, such as detecting the trigger conditions of each sensor in sequence from the direction where the train approaches to the direction where it leaves, so as to accurately calculate the train speed and position. Reordering may involve updating the index values or address information of the sensors in the software algorithm of the system to ensure that the system can correctly identify the position and order of each sensor.

[0104] Finally, reset the distance between adjacent optoelectronic sensors.

[0105] After shielding the faulty sensor and reordering, the distance between adjacent optoelectronic sensors has changed. The original distance parameter was set based on all sensors operating normally. Now, due to the removal of a certain sensor, the distance between the new adjacent sensors is different from before. If the distance parameter is not reset, the algorithm for calculating data such as train speed based on the sensor spacing will generate errors, resulting in inaccurate speed measurement, which in turn affects the prediction accuracy of the time when the train reaches the preset acquisition position and the accuracy of the entire image acquisition system.

[0106] It should be noted that in this article, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitations, an element defined by the statement "including a..." does not exclude the existence of another identical element in the process, method, article or device including that element.

[0107] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions for causing a computer terminal (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in various embodiments of the present invention.

[0108] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the spirit and scope protected by the claims of the present invention. All of these are within the protection scope of the present invention.

Claims

1. A train image acquisition system, characterized in that include: Speed ​​measuring subsystem, used to measure the speed of the train; It includes a high-speed detection sensor group and a low-speed detection sensor group; The high-speed detection sensor group is used to detect the speed of a train whose speed is higher than the speed threshold, and the low-speed detection sensor group is used to detect the speed of a train whose speed is less than or equal to the speed threshold; A controller, used to predict the arrival time of the train at the preset acquisition position using the train speed obtained by the speed measurement subsystem, and calculate the image acquisition trigger time of the image acquisition module according to the arrival time; Sending an instruction to the shooting module to perform image acquisition at the image acquisition trigger time; The image acquisition module is used to acquire images in response to instructions sent by the controller.

2. A train photography system according to claim 1, characterized in that: The high-speed detection sensor group includes wheel sensors arranged along the track front and back; it also includes a data acquisition box, which calculates the speed of the train based on the distance between the front and rear wheel sensors and the time difference between the front and rear wheel sensors detecting the same wheel.

3. A train photography system according to claim 1, characterized in that: The low-speed detection sensor group includes a sensor array consisting of at least two photoelectric sensors; the photoelectric sensor includes a transmitting end and a receiving end respectively arranged on both sides of the track; the light emitted by several transmitting ends in the sensor array forms a light curtain for detecting the wheel travel speed.

4. A train photography system according to claim 1, characterized in that: The low speed detection sensor group detects the train and determines the running speed range of the train.

5. A camera precision triggering method based on vehicle speed prediction, applied to the train image acquisition system according to any one of claims 1 to 4, characterized in that include: Determine whether a train is approaching, and determine the direction and speed range of the train; If the train speed is greater than the set speed threshold, the high-speed detection sensor group is used to detect the speed of the train, otherwise the low-speed detection sensor group is used to detect the speed of the train; Using the detected train speed to predict the time when the train will arrive at the preset collection location; The image acquisition module is triggered to acquire the train wheel image according to the arrival time.

6. The camera accurate triggering method based on vehicle speed prediction according to claim 5 is characterized in that The method for the high-speed detection sensor group to detect the speed of a train includes: Determine whether the wheel sensor forward relative to the train direction is triggered; if triggered, determine the validity of the trigger signal; When the trigger signal is valid, determine whether the speed measurement process has timed out; if not, determine whether the wheel sensor at the rear relative to the train direction has been triggered; if triggered, determine the validity of the trigger signal; When the trigger signal is valid, the current speed of the train is calculated based on the distance between the two wheel sensors and the time difference between the front and rear wheel sensors being triggered by the same wheel.

7. The camera accurate triggering method based on vehicle speed prediction according to claim 5 is characterized in that The method for the low speed detection sensor group to detect the speed of the train includes: From the train direction to the outgoing direction, determine whether the photoelectric sensor is triggered. If triggered, determine the validity of the trigger signal; When the trigger signal is valid, the current speed of the train is calculated according to the distance between any two photoelectric sensors and the time difference between the two photoelectric sensors being triggered by the same wheel.

8. A camera precise triggering method based on vehicle speed prediction according to claim 6 or 7, characterized in that: After calculating the current speed of the train, determine whether the speed meets the train speed trend; If not satisfied, the current speed is corrected.

9. The camera precision triggering method based on vehicle speed prediction according to claim 5, characterized in that Methods for predicting the time when a train arrives at a preset collection location include: When the current train speed is higher than the speed threshold, the delay of the wheel sensor is obtained according to the current train speed; The theoretical time for the train to reach the preset collection position is calculated using the formula t1=s1 / v, where t1 is the theoretical time, s1 is the distance between the wheel sensor at the rear relative to the train direction and the preset collection position, and v is the current speed of the train; The time it takes for the train to arrive at the preset collection location is calculated using the formula t=t1-t2, where t is the arrival time, t1 is the theoretical time, and t2 is the delay time of the wheel sensor; When the current train speed is lower than or equal to the speed threshold, the delay without the photoelectric sensor is obtained according to the current train speed; The theoretical time for the train to reach the preset collection position is calculated using the formula t1=s2 / v, where t1 is the theoretical time, s2 is the distance between the last triggered photoelectric sensor and the preset collection position, and v is the current speed of the train; The time it takes for the train to arrive at the preset collection location is calculated using the formula t=t1-t3, where t is the arrival time, t1 is the theoretical time, and t3 is the delay time of the photoelectric sensor.

10. The camera precise triggering method based on vehicle speed prediction according to claim 5, characterized in that: Check if the photoelectric sensor is faulty; In the event that a photoelectric sensor fails, the photoelectric sensor is shielded and all photoelectric sensors are reordered; Reset the distance between adjacent photoelectric sensors.

Images