A railway station liaison operation monitoring method and system based on intelligent identification

Abstract

The present invention relates to the technical field of operation monitoring, and specifically discloses a railway station liaison operation monitoring method and system based on intelligent identification, comprising the following steps: step S1: selecting a sample of station liaison officers, and collecting action video samples of the station liaison officers when using communication equipment; extracting common limb movement features of all samples through a limb analysis model; collecting facial images of the samples when performing tasks, and calculating the blinking frequency of each sample; generating a unified frequency reference value based on all blinking frequencies; step S2: determining the liaison monitoring duration, and after real-time monitoring of key moments of train status changes, defining a liaison monitoring time period based on the monitoring duration; obtaining real-time video data of the station liaison officers, extracting their limb movement features through a limb analysis model, and calculating the similarity with the common features; collecting facial images of the station liaison officers in real time, and analyzing the current blinking frequency; and step S3: generating a work status index by combining a limb index and a concentration index.

Description

Technical Field

[0001] The present invention relates to the technical field of operation monitoring, and in particular to a method and system for monitoring railway station communication operations based on intelligent identification. Background Art

[0002] Railway station liaison operation monitoring is an important measure taken during railway construction and maintenance operations to ensure the safety of operators and safeguard railway transportation order. It ensures the safety of operators and smooth railway transportation through multiple monitoring methods and strict monitoring processes.

[0003] Railway liaison operations require meticulous follow-up at every stage, necessitating a high level of staff focus. However, existing monitoring technologies primarily focus on the work status of on-site personnel, but are insufficient for monitoring the work status of station liaison officers. This can not only lead to decreased work efficiency but also potentially cause safety incidents due to inattention. Therefore, developing technology that can effectively monitor the work status of station liaison officers is crucial. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for monitoring railway station liaison operations based on intelligent identification to solve the above technical problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A method for monitoring railway station liaison operations based on intelligent identification comprises the following steps:

[0007] Step S1: Selecting a number of on-site liaison officer samples, obtaining action video samples of the on-site liaison officer samples when using communication equipment to make contacts, and establishing a body movement analysis model. Using the body movement analysis model, obtain the common body movement features of the on-site liaison officer samples in all action video samples;

[0008] Setting a work task, obtaining facial image samples of each resident liaison officer sample when performing the work task, obtaining the blinking frequency of each resident liaison officer sample based on the facial image samples, and obtaining a frequency reference value based on all blinking frequencies;

[0009] Step S2: Obtain the duration of all action video samples, and set the contact monitoring duration according to the duration; obtain the status of the train in real time, the status including parking and moving; record the time when the status of the train changes as the change time, and select the contact monitoring time period after the change time according to the contact monitoring duration;

[0010] Obtain video data of the station liaison officer during the contact monitoring period, input the video data into the body analysis model, obtain the body movement features of the station liaison officer in the video data; and obtain the similarity between the body movement features and the public body movement features,

[0011] Obtaining a workbench of the resident liaison officer, setting a monitoring point on the workbench, wherein the monitoring point is used to obtain a facial image of the resident liaison officer in real time; obtaining a blinking frequency of the resident liaison officer based on the facial image;

[0012] Step S3: Set the limb index I1. If the similarity is greater than or equal to a preset similarity threshold, record the limb index I1=1; otherwise, record the limb index I1=0; set the concentration index I2. If the blinking frequency exceeds the frequency reference value, record the concentration index I2=1; otherwise, record the concentration index I2=0; obtain the work status index based on the concentration index and the limb index. If and only if the work status index WI=1, there is no abnormality in the work status of the on-site liaison officer.

[0013] As a further solution of the present invention: the acquisition of the action video samples is based on a motion capture device, and the action video samples include action videos of the on-site liaison samples when using communication equipment, acquired from different angles.

[0014] As a further solution of the present invention: the process of obtaining the blinking frequency of the resident liaison officer sample includes:

[0015] Establish an image recognition model to identify all facial image samples. Number all facial image samples according to the time series, and perform image recognition on the facial image samples through the image recognition model to obtain recognition results, which include closed eyes and open eyes; obtain the numbers of facial image samples whose recognition results are closed eyes {n1, n2, ..., n m}, where n m Indicates the number of the mth facial image sample with closed eyes as the recognition result, and m is the total number of facial image samples with closed eyes as the recognition result; get the blink frequency , where n i It represents the number of the facial image sample whose recognition result is closed eyes, i∈[1,m-1] and i is a positive integer, and t is the unit time.

[0016] As a further solution of the present invention: the contact monitoring duration is an average duration of all action video samples.

[0017] As a further solution of the present invention: when the working status index WI=0, the working status of the on-site liaison officer is abnormal.

[0018] As a further solution of the present invention: the process of establishing the limb analysis model includes:

[0019] The action video sample is divided into several image frames, and the limb parts of the resident liaison officer sample in the image frames are labeled and segmented based on computer vision technology, the joints and limb contours are marked, the joints and contours are recorded as limb information, and the action names corresponding to the limb information are marked to obtain sample data; an initial model is established based on a convolutional neural network, and the initial model is trained using the sample data to obtain a limb analysis model.

[0020] As a further solution of the present invention: the process of obtaining the frequency reference value includes:

[0021] Get the average of all blink frequencies, recorded as Bf ave , and obtain the standard deviation s of all blink frequencies, then the frequency reference value Frv=Bf ave -z×s, where z is the multiple threshold and z∈[0, 3].

[0022] A railway station liaison operation monitoring system based on intelligent identification, comprising:

[0023] Sample collection module: selects several samples of resident liaison officers, obtains action video samples of the resident liaison officers when using communication equipment to make contacts, and establishes a body analysis model. Through the body analysis model, the common body movement features of the resident liaison officers in all action video samples are obtained;

[0024] Setting a work task, obtaining facial image samples of each resident liaison officer sample when performing the work task, obtaining the blinking frequency of each resident liaison officer sample based on the facial image samples, and obtaining a frequency reference value based on all blinking frequencies;

[0025] Monitoring module: obtains the duration of all action video samples and sets the contact monitoring duration according to the duration; obtains the status of the train in real time, including parking and driving; records the time when the status of the train changes as the change time, and selects the contact monitoring time period after the change time according to the contact monitoring duration;

[0026] Obtain video data of the station liaison officer during the contact monitoring period, input the video data into the body analysis model, obtain the body movement features of the station liaison officer in the video data; and obtain the similarity between the body movement features and the public body movement features,

[0027] Obtaining a workbench of the resident liaison officer, setting a monitoring point on the workbench, wherein the monitoring point is used to obtain a facial image of the resident liaison officer in real time; obtaining a blinking frequency of the resident liaison officer based on the facial image;

[0028] Judgment module: Set the limb index I1. If the similarity is greater than or equal to a preset similarity threshold, record the limb index I1=1; otherwise, record the limb index I1=0; set the concentration index I2. If the blinking frequency exceeds the frequency reference value, record the concentration index I2=1; otherwise, record the concentration index I2=0; obtain the work status index based on the concentration index and the limb index. If and only if the work status index WI=1, the work status of the resident liaison officer is normal.

[0029] Beneficial effects of the present invention:

[0030] By establishing a limb analysis model and a blink frequency baseline value, the present invention can more accurately evaluate the working status of the station liaison officer, thereby reducing false alarms and missed alarms; the method can obtain the train status and the video data of the station liaison officer in real time, and make timely judgments and adjustments to the working status, which helps to improve response speed and work efficiency; through the automatic analysis of video data and facial images by intelligent algorithms, the need for manual intervention is reduced and the automation level of the monitoring process is improved; through continuous monitoring of the working status of the station liaison officer, irregular movements or inattention can be discovered and corrected in time, which helps to prevent accidents and ensure the safety of railway transportation; the collected data can be used to further analyze and optimize the work process, provide decision support for management, and promote the improvement and optimization of the operating process. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The present invention will be further described below with reference to the accompanying drawings.

[0032] Figure 1 The present invention is a flow chart of a method for monitoring railway station liaison operations based on intelligent identification. DETAILED DESCRIPTION

[0033] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0034] See also Figure 1 As shown, the present invention is a method for monitoring railway station liaison operations based on intelligent identification, comprising the following steps:

[0035] Step S1: Selecting a number of on-site liaison officer samples, obtaining action video samples of the on-site liaison officer samples when using communication equipment to make contacts, and establishing a body movement analysis model. Using the body movement analysis model, obtain the common body movement features of the on-site liaison officer samples in all action video samples;

[0036] Setting a work task, obtaining facial image samples of each resident liaison officer sample when performing the work task, obtaining the blinking frequency of each resident liaison officer sample based on the facial image samples, and obtaining a frequency reference value based on all blinking frequencies;

[0037] It is understandable that the core of the above steps is to provide action norms and physiological benchmarks for subsequent monitoring through data-driven standardized modeling; select multiple groups of on-site liaison officers, record their action video samples when using communication equipment, use computer vision technology (such as posture estimation and action segmentation) to analyze the action video samples, and extract key action nodes (such as gestures, body posture, and equipment operation procedures); use machine learning (such as cluster analysis and feature dimensionality reduction) to extract common action patterns from multiple samples to form a standardized public body action feature library; eliminate individual differences and establish standardized action templates for subsequent real-time action compliance comparison;

[0038] At the same time, when the station liaison officer performs a specific task, the camera collects facial image samples, uses image recognition algorithms (such as Haar cascade and deep learning eye segmentation) to detect the open and closed state of the eyes, and counts the time consumed by a single blink. The blink frequency of all samples is statistically analyzed to generate a frequency baseline value.

[0039] It should be noted that combining motion and physiological indicators (limbs + blinking) avoids the limitations of a single indicator; establishing a benchmark through sample data adapts to different scenarios and individual differences, improves the robustness of the monitoring system, and transforms manual experience into a data model, achieving an upgrade from subjective judgment to objective intelligent analysis;

[0040] As a preferred embodiment of the present invention, the acquisition of the action video samples is based on a motion capture device, and the action video samples include action videos of the station liaison officer samples using the communication equipment acquired from different angles;

[0041] It is understandable that multiple cameras or sensors are arranged around the station liaison to simultaneously collect action videos from different angles. The data is integrated through multi-view fusion algorithms (such as 3D reconstruction and feature matching) to generate a complete action model; ensuring the comprehensiveness of the action data, avoiding the loss of action information caused by a single viewpoint, improving the robustness of the action analysis, and adapting to different scenes and posture changes;

[0042] As a preferred embodiment of the present invention, the process of establishing the limb analysis model includes:

[0043] Dividing the action video sample into a number of image frames, annotating and segmenting the limbs of the resident liaison officer sample in the image frames based on computer vision technology, marking the joints and limb contours, recording the joints and contours as limb information, and marking the action names corresponding to the limb information to obtain sample data; establishing an initial model based on a convolutional neural network, and training the initial model using the sample data to obtain a limb analysis model;

[0044] It should be noted that computer vision technology is used to annotate and segment the limb parts in the image frames, marking the joints (such as elbows and knees) and limb contours (such as arms and legs) to form structured limb information; each action is annotated with the corresponding action name (such as raising a hand, turning around) to generate labeled sample data;

[0045] As a preferred embodiment of the present invention, the process of obtaining the blinking frequency of the resident liaison officer sample includes:

[0046] Establish an image recognition model to identify all facial image samples. Number all facial image samples according to the time series, and perform image recognition on the facial image samples through the image recognition model to obtain recognition results, which include closed eyes and open eyes; obtain the numbers of facial image samples whose recognition results are closed eyes {n1, n2, ..., n m}, where n m Indicates the number of the mth facial image sample with closed eyes as the recognition result, and m is the total number of facial image samples with closed eyes as the recognition result; get the blink frequency , where n i represents the number of the i-th facial image sample whose recognition result is closed eyes, i∈[1,m-1] and i is a positive integer, t is the unit time;

[0047] It should be noted that the time interval between adjacent closed-eye images is calculated. For example, the number of the i-th closed-eye image is n. i , the number of the i+1th closed-eye image is n i+1 , then the time interval between the two is n i+1 -n i , represents the total time span from the first eye closure to the last eye closure; m-1 represents the number of eye closures minus 1 (i.e., the number of blinks); the unit time t is used to standardize the frequency value (such as the number of blinks per minute);

[0048] It is worth noting that blinking frequency is an important physiological indicator for measuring concentration, because it can directly reflect a person's attention state and fatigue level. When a person is in a daze or sleepy, the blinking frequency is usually significantly lower than the frequency baseline value; when in a daze, the human brain is in a relaxed or unconscious state, and the eye muscle activity is reduced, resulting in a lower blinking frequency. At this time, when the attention is distracted or there is a lack of external stimulation, the brain's control over eye movement is weakened, and the blinking frequency naturally decreases; when you are sleepy, fatigue will cause the tension of the eye muscles to decrease, the eyelid closure time to prolong, and thus reduce the number of blinks. At this time, as the sense of fatigue increases, the brain's alertness decreases, and the ability to regulate eye movement is weakened, resulting in a decrease in blinking frequency;

[0049] In a preferred embodiment of the present invention, the process of obtaining the frequency reference value includes:

[0050] Get the average of all blink frequencies, recorded as Bf ave , and obtain the standard deviation s of all blink frequencies, then the frequency reference value Frv=Bf ave -z×s, where z is the multiple threshold and z∈[0,3];

[0051] It is understandable that different people have different baseline blink rates (such as those who are born with high or low blink rates). By combining the mean and standard deviation with the multiple threshold, the baseline value can be dynamically adjusted to make it more consistent with the actual physiological characteristics of the individual or group. This can avoid misjudgments caused by individual differences and improve the accuracy and fairness of the monitoring system. By adjusting the multiple threshold, for example, the smaller the z value, the closer the baseline value is to the mean, and the monitoring is more stringent; the larger the z value, the looser the baseline value is, and more individual differences can be accommodated.

[0052] Step S2: Obtain the duration of all action video samples, and set the contact monitoring duration according to the duration; obtain the status of the train in real time, the status including parking and moving; record the time when the status of the train changes as the change time, and select the contact monitoring time period after the change time according to the contact monitoring duration;

[0053] Obtain video data of the station liaison officer during the contact monitoring period, input the video data into the body analysis model, obtain the body movement features of the station liaison officer in the video data; and obtain the similarity between the body movement features and the public body movement features,

[0054] Obtaining a workbench of the resident liaison officer, setting a monitoring point on the workbench, wherein the monitoring point is used to obtain a facial image of the resident liaison officer in real time; obtaining a blinking frequency of the resident liaison officer based on the facial image;

[0055] It should be noted that there are more than one contact time periods, but several within the entire operating time period. The train state change moment is a moment when the train changes from static to dynamic, or from dynamic to static, and the contact time period is a period of time before or after the change moment, so as to monitor whether the station liaison officer has made timely contact during the contact time period.

[0056] It is understandable that a reasonable contact monitoring duration is determined based on the duration of all action video samples; contact monitoring time periods are delineated according to the critical moments of train status changes (such as stopping and moving), ensuring monitoring during the critical period; avoiding waste of resources for round-the-clock monitoring, focusing on high-risk periods, and improving monitoring efficiency; train status is divided into two types: stopping and moving, and the moment of status change (such as stopping and moving) is marked as the change moment; after the change moment, the monitoring task is started according to the set contact monitoring duration; ensuring real-time monitoring of the work status of the station liaison officer during the critical period of train status changes (such as starting or stopping); by evaluating whether the station liaison officer uses communication equipment at the moment of train status change;

[0057] As a preferred embodiment of the present invention, the contact monitoring duration is an average duration of all action video samples;

[0058] Step S3: Setting a limb index I1. If the similarity is greater than or equal to a preset similarity threshold, the limb index I1 is recorded as 1; otherwise, the limb index I1 is recorded as 0. Setting a focus index I2. If the blinking frequency exceeds the frequency reference value, the focus index I2 is recorded as 1; otherwise, the focus index I2 is recorded as 0. Based on the focus index and the limb index, a work status index is obtained. If and only if the work status index WI is 1, the work status of the station liaison officer is normal.

[0059] It should be noted that the limb index is used to assess the standardization of the resident liaison officer's movements. If the similarity between the real-time movement characteristics and the public limb movement characteristics is greater than or equal to a preset threshold, then I1 = 1 (normal movement); otherwise, I2 = 0 (abnormal movement). The similarity between the real-time movement characteristics and the standard movement template is calculated using a limb analysis model. The similarity threshold is determined based on historical data or experiments to ensure the accuracy of the judgment. Quantifying movement compliance provides a basis for work status assessment.

[0060] It should be noted that the focus index is used to assess the focus of on-site liaison officers. If the real-time blink frequency is ≤ the frequency baseline value, then I2 = 1 (focused); otherwise, I2 = 0 (distracted or fatigued). The blink frequency is calculated using image recognition technology and compared with the frequency baseline value. The frequency baseline value is dynamically adjusted based on group data to accommodate individual differences. This quantifies focus and provides a supplementary indicator for work status assessment.

[0061] As a preferred embodiment of the present invention, when the working status index WI=0, the working status of the station liaison officer is abnormal;

[0062] It should be noted that the work status index is the logical AND result of the limb index and the concentration index. It integrates the limb index and the concentration index through logical operations to generate a comprehensive evaluation result, supporting real-time monitoring and dynamic early warning; it comprehensively evaluates the work status through multi-dimensional indicators (action + concentration) to improve the accuracy and reliability of judgment.

[0063] It is worth noting that the present invention also includes intelligent judgment of the on-site situation in the station, and the process of intelligent judgment includes:

[0064] 1. Intelligent identification and notification of incoming vehicle information:

[0065] Oncoming train detection: Real-time acquisition of train location and operating status through sensors, track circuits, or train positioning systems; use of artificial intelligence algorithms to predict train arrival times and ensure early warning;

[0066] Automatic notification: When an approaching train is detected, a notification is automatically sent to the station liaison officer, who is required to respond promptly and notify the on-site operator via communication equipment;

[0067] Action confirmation: Through video surveillance or motion capture technology, the station liaison officer is monitored in real time to see if he or she has executed the notification action. If there is no timely response, an early warning is triggered to remind the liaison officer to perform his or her duties.

[0068] 2. Intelligent verification of personnel leaving the road:

[0069] Satellite positioning and video surveillance: Using satellite positioning devices worn by on-site workers, their location information is obtained in real time. Combined with on-site video surveillance, this allows confirmation that workers have left the road and entered a safe area as required.

[0070] Distance calculation and safety assessment: Calculate the real-time distance between the personnel location and the track to determine whether safety standards are met; utilize geographic information systems (GIS) and spatial analysis algorithms to dynamically define safety zone boundaries to ensure accuracy of assessments;

[0071] Intelligent verification and early warning: If personnel fail to leave the road or enter the safe zone as required, an automatic early warning will be triggered, notifying the station liaison and on-site management personnel. This system also optimizes the demarcation and judgment logic of safe zones by combining historical data with machine learning models, enhancing the system's intelligence.

[0072] 3. Data integration and comprehensive analysis:

[0073] Multi-source data fusion: Integrate multi-source data such as train location, personnel positioning, and video surveillance to build a comprehensive on-site operation status map. Through big data analysis and visualization technology, the status of trains, personnel, and equipment is displayed in real time to assist decision-making.

[0074] Anomaly detection and processing: Utilize anomaly detection algorithms to identify behaviors that do not comply with safety regulations; automatically generate processing suggestions and distribute them to relevant personnel through the system;

[0075] A railway station liaison operation monitoring system based on intelligent identification, comprising:

[0076] Sample collection module: selects several samples of resident liaison officers, obtains action video samples of the resident liaison officers when using communication equipment to make contacts, and establishes a body analysis model. Through the body analysis model, the common body movement features of the resident liaison officers in all action video samples are obtained;

[0077] Setting a work task, obtaining facial image samples of each resident liaison officer sample when performing the work task, obtaining the blinking frequency of each resident liaison officer sample based on the facial image samples, and obtaining a frequency reference value based on all blinking frequencies;

[0078] Monitoring module: obtains the duration of all action video samples and sets the contact monitoring duration according to the duration; obtains the status of the train in real time, including parking and driving; records the time when the status of the train changes as the change time, and selects the contact monitoring time period after the change time according to the contact monitoring duration;

[0079] Obtain video data of the station liaison officer during the contact monitoring period, input the video data into the body analysis model, obtain the body movement features of the station liaison officer in the video data; and obtain the similarity between the body movement features and the public body movement features,

[0080] Obtaining a workbench of the resident liaison officer, setting a monitoring point on the workbench, wherein the monitoring point is used to obtain a facial image of the resident liaison officer in real time; obtaining a blinking frequency of the resident liaison officer based on the facial image;

[0081] Judgment module: Set the limb index I1. If the similarity is greater than or equal to a preset similarity threshold, record the limb index I1=1; otherwise, record the limb index I1=0; set the concentration index I2. If the blinking frequency exceeds the frequency reference value, record the concentration index I2=1; otherwise, record the concentration index I2=0; obtain the work status index based on the concentration index and the limb index. If and only if the work status index WI=1, the work status of the resident liaison officer is normal.

[0082] The above is a detailed description of an embodiment of the present invention. However, the content described is only a preferred embodiment of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for monitoring railway station liaison operations based on intelligent identification, characterized in that: The following steps are involved: Step S1: Selecting a number of on-site liaison officer samples, obtaining action video samples of the on-site liaison officer samples when using communication equipment to make contacts, and establishing a body movement analysis model. Using the body movement analysis model, obtain the common body movement features of the on-site liaison officer samples in all action video samples; Setting a work task, obtaining facial image samples of each resident liaison officer sample when performing the work task, obtaining the blinking frequency of each resident liaison officer sample based on the facial image samples, and obtaining a frequency reference value based on all blinking frequencies; Step S2: Obtain the duration of all action video samples, and set the contact monitoring duration according to the duration; obtain the status of the train in real time, the status including parking and moving; record the time when the status of the train changes as the change time, and select the contact monitoring time period after the change time according to the contact monitoring duration; Obtain video data of the station liaison officer during the contact monitoring period, input the video data into the body analysis model, obtain the body movement features of the station liaison officer in the video data; and obtain the similarity between the body movement features and the public body movement features, Obtaining a workbench of the resident liaison officer, setting a monitoring point on the workbench, wherein the monitoring point is used to obtain a facial image of the resident liaison officer in real time; obtaining a blinking frequency of the resident liaison officer based on the facial image; Step S3: Set the limb index I1. If the similarity is greater than or equal to a preset similarity threshold, record the limb index I1=1; otherwise, record the limb index I1=0; set the concentration index I2. If the blinking frequency exceeds the frequency reference value, record the concentration index I2=1; otherwise, record the concentration index I2=0; obtain the work status index based on the concentration index and the limb index. If and only if the work status index WI=1, there is no abnormality in the work status of the on-site liaison officer.

2. The method for monitoring railway station liaison operations based on intelligent identification according to claim 1, characterized in that: In step S1, the acquisition of the action video samples is based on a motion capture device, and the action video samples include action videos of the on-site liaison officer samples using communication equipment acquired from different angles.

3. The method for monitoring railway station liaison operations based on intelligent identification according to claim 1, characterized in that: In step S1, the process of establishing the limb analysis model includes: The action video sample is divided into several image frames, and the limb parts of the resident liaison officer sample in the image frames are labeled and segmented based on computer vision technology, the joints and limb contours are marked, the joints and contours are recorded as limb information, and the action names corresponding to the limb information are marked to obtain sample data; an initial model is established based on a convolutional neural network, and the initial model is trained using the sample data to obtain a limb analysis model.

4. The method for monitoring railway station liaison operations based on intelligent identification according to claim 1, characterized in that: In step S1, the process of obtaining the blinking frequency of the resident liaison officer sample includes: Establish an image recognition model to identify all facial image samples. Number all facial image samples according to the time series, and perform image recognition on the facial image samples through the image recognition model to obtain recognition results, which include closed eyes and open eyes; obtain the numbers of facial image samples whose recognition results are closed eyes {n1, n2, ..., n m }, where n m Indicates the number of the mth facial image sample with closed eyes as the recognition result, and m is the total number of facial image samples with closed eyes as the recognition result; get the blink frequency , where n i It represents the number of the facial image sample whose recognition result is closed eyes, i∈[1,m-1] and i is a positive integer, and t is the unit time.

5. The method for monitoring railway station liaison operations based on intelligent identification according to claim 1, characterized in that: In step S1, the process of obtaining the frequency reference value includes: Get the average of all blink frequencies, recorded as Bf ave , and obtain the standard deviation s of all blink frequencies, then the frequency reference value Frv=Bf ave -z×s, where z is the multiple threshold and z∈[0, 3].

6. The method for monitoring railway station liaison operations based on intelligent identification according to claim 1, characterized in that: In step S2, the contact monitoring duration is the average duration of all action video samples.

7. The method for monitoring railway station liaison operations based on intelligent identification according to claim 1, characterized in that: In step S3, when the working status index WI=0, the working status of the on-site liaison officer is abnormal.

8. A railway station liaison operation monitoring system based on intelligent identification, characterized in that: include: Sample collection module: selects several samples of resident liaison officers, obtains action video samples of the resident liaison officers when using communication equipment to make contacts, and establishes a body analysis model. Through the body analysis model, the common body movement features of the resident liaison officers in all action video samples are obtained; Setting a work task, obtaining facial image samples of each resident liaison officer sample when performing the work task, obtaining the blinking frequency of each resident liaison officer sample based on the facial image samples, and obtaining a frequency reference value based on all blinking frequencies; Monitoring module: obtains the duration of all action video samples and sets the contact monitoring duration according to the duration; obtains the status of the train in real time, including parking and driving; records the time when the status of the train changes as the change time, and selects the contact monitoring time period after the change time according to the contact monitoring duration; Obtain video data of the station liaison officer during the contact monitoring period, input the video data into the body analysis model, obtain the body movement features of the station liaison officer in the video data; and obtain the similarity between the body movement features and the public body movement features, Obtaining a workbench of the resident liaison officer, setting a monitoring point on the workbench, wherein the monitoring point is used to obtain a facial image of the resident liaison officer in real time; obtaining a blinking frequency of the resident liaison officer based on the facial image; Judgment module: Set the limb index I1. If the similarity is greater than or equal to a preset similarity threshold, record the limb index I1=1; otherwise, record the limb index I1=0; set the concentration index I2. If the blinking frequency exceeds the frequency reference value, record the concentration index I2=1; otherwise, record the concentration index I2=0; obtain the work status index based on the concentration index and the limb index. If and only if the work status index WI=1, the work status of the resident liaison officer is normal.

Images