Intelligent rail-mounted area safety early warning method and device based on edge calculation

Abstract

The invention discloses a smart rail-mounted area safety early warning method and device based on edge computing, and relates to the technical field of safety early warning, and the method comprises the steps: collecting a video stream, a three-dimensional vibration spectrum and a microenvironment parameter set in a coverage area in real time, constructing a time-space associated multi-modal data cube according to the video stream, the three-dimensional vibration spectrum and the microenvironment parameter set, extracting depth features of each modal in the multi-modal data cube, carrying out dynamic weighting on the depth features, generating a risk feature tensor with time-space consistency, determining a neighborhood risk factor based on the risk feature tensor, and obtaining a risk result; determining a risk positioning coordinate and a dynamic risk level based on the neighborhood risk factor, performing post-quantum encryption on the risk positioning coordinate and the dynamic risk level to obtain a risk fingerprint, generating an alarm message based on the risk fingerprint, embedding the alarm message into a guarantee time window, and uploading the alarm message to a central cloud platform, so that the central cloud platform triggers an early warning response. Accurate, real-time and safe perception and response to the risk are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of safety early warning, in particular to a smart rail area safety early warning method and device based on edge computing. BACKGROUND

[0002] With the development of rail transit towards high speed and high density, the safe operation and maintenance of rail areas (including tracks, tunnels, stations, etc.) are facing unprecedented challenges. Existing systems mostly use isolated monitoring methods, but false positives or false negatives often occur, so the existing rail area safety early warning technology cannot meet the safety requirements.

[0003] The above content is only used to assist in understanding the technical solutions of the present application and does not represent the acknowledgement of the above content as prior art. SUMMARY

[0004] The main purpose of the present application is to provide a smart rail area safety early warning method and device based on edge computing, aiming to solve the technical problem of low early warning efficiency caused by data isolation in the existing rail area safety monitoring.

[0005] To achieve the above purpose, the present application provides a smart rail area safety early warning method based on edge computing, the method comprising: collecting video streams, three-dimensional vibration spectra and micro-environment parameter sets in a covered area in real time, and constructing a multi-modal data cube with spatio-temporal correlation according to the video streams, the three-dimensional vibration spectra and the micro-environment parameter sets; extracting deep features of each modality in the multi-modal data cube, dynamically weighting the deep features, and generating a risk feature tensor with spatio-temporal consistency; determining a neighborhood risk factor based on the risk feature tensor, determining a risk positioning coordinate and a dynamic risk level based on the neighborhood risk factor; post-quantum encrypting the risk positioning coordinate and the dynamic risk level to obtain a risk fingerprint, and generating an alarm message based on the risk fingerprint; embedding the alarm message into a protection time window and uploading it to a central cloud platform, so that the central cloud platform triggers an early warning response.

[0006] In an embodiment, the step of collecting video streams, three-dimensional vibration spectra and micro-environment parameter sets in a covered area in real time, and constructing a multi-modal data cube with spatio-temporal correlation according to the video streams, the three-dimensional vibration spectra and the micro-environment parameter sets comprises: aligning the clock information of all sensors in the covered area, establishing a global timestamp according to the clock information, mapping the position information of all sensors to a three-dimensional space coordinate system, and generating a spatio-temporal unified network according to the global timestamp and the three-dimensional space coordinate system; dynamically modeling a background and segmenting a foreground of a video stream covering a region, determining a moving target in the region, and determining a spatio-temporal dynamic feature of the moving target; wavelet packet decomposing a three-dimensional vibration spectrum of a region covered, extracting an impact energy feature, and constructing an abnormal pattern recognition matrix based on the impact energy feature; fusing a micro-environment parameter set through a Kalman filter to generate an environment risk index; in the spatio-temporal unified network, constructing a spatio-temporally correlated multi-modal data cube based on the spatio-temporal dynamic feature, the abnormal pattern recognition matrix, and the environment risk index.

[0007] In an embodiment, the step of dynamically modeling a background and segmenting a foreground of a video stream covering a region, determining a moving target in the region, and determining a spatio-temporal dynamic feature of the moving target comprises: determining an illumination intensity in the video stream covering a region, generating a dynamic light intensity weight based on the illumination intensity, and creating a dynamic background based on the dynamic light intensity weight; performing a morphological top-hat transformation on the video stream, determining a contrast of small targets in the video stream, determining a foreground mask based on the contrast, segmenting a foreground based on the foreground mask, and determining a foreground target; projecting the foreground target to the dynamic background to determine a moving target in the region; calculating a minimum circumscribed rectangle of the moving target based on the foreground mask, extracting a feature vector of the minimum circumscribed rectangle, and generating a spatio-temporal dynamic feature based on the feature vector.

[0008] In an embodiment, the step of wavelet packet decomposing a three-dimensional vibration spectrum of a region covered, extracting an impact energy feature, and constructing an abnormal pattern recognition matrix comprises: performing adaptive noise complete ensemble empirical mode decomposition on a three-dimensional vibration spectrum of a region covered to generate an enhanced signal; wavelet packet decomposing the enhanced signal to obtain a plurality of sub-bands; calculating an energy entropy of each of the sub-bands, and marking a sub-band of a preset frequency in the energy entropy as an impact sensitive frequency band; determining an energy mutation rate of the impact sensitive frequency band, marking the impact sensitive frequency band as an abnormal pattern when the energy mutation rate is greater than an abnormal mutation rate, and extracting an impact energy feature of the impact sensitive frequency band; determining a time feature, a frequency domain feature, and a spatial feature of the impact energy feature, and generating an abnormal pattern recognition matrix based on the time feature, the frequency domain feature, and the spatial feature.

[0009] In an embodiment, the step of fusing the micro-environment parameter set by the Kalman filter to generate the environment risk index comprises: interpolating and compensating the micro-environment parameter set to obtain a micro-environment parameter pre-processing set; determining a state vector according to the micro-environment parameter pre-processing set, and fusing the state vector by an observation equation of the Kalman filter to obtain a micro-environment filtering parameter; generating an environment risk index according to an environment risk weight factor and the micro-environment filtering parameter.

[0010] In an embodiment, the step of extracting the deep features of each modality in the multi-modal data cube, dynamically weighting the deep features to generate a risk feature tensor with spatio-temporal consistency comprises: extracting spatial features of a video stream and spatio-temporal features of a vibration spectrum in the multi-modal data cube by a double-path convolution; aligning the spatial features and the spatio-temporal features to obtain spatio-temporally aligned deep features; determining an environment risk index, and generating an attention weight according to the environment risk index; dynamically weighting the deep features based on the attention weight to generate a risk feature tensor with spatio-temporal consistency.

[0011] In an embodiment, the step of determining a neighborhood risk factor based on the risk feature tensor, determining a risk positioning coordinate and a dynamic risk level based on the neighborhood risk factor comprises: determining a physical connection relationship of each device in the coverage area, and determining a device topology relationship based on the physical connection relationship; determining a neighborhood relationship of each device based on the device topology relationship; spatio-temporally convolving the neighborhood relationship to aggregate neighborhood risk information of a target hop count; determining a neighborhood risk factor and a risk gradient according to the neighborhood risk information; determining a risk positioning coordinate according to the risk gradient, and determining a dynamic risk level according to the neighborhood risk factor and the risk positioning coordinate.

[0012] In an embodiment, the step of post-quantum encrypting the risk positioning coordinate and the dynamic risk level to obtain a risk fingerprint, and generating an alarm message based on the risk fingerprint comprises: post-quantum encrypting the risk positioning coordinate and the dynamic risk level to obtain encrypted information; performing a hash operation on the encrypted information to obtain a risk fingerprint; packaging the risk fingerprint and the encrypted information to obtain an alarm message.

[0013] In an embodiment, the step of embedding the alarm message into a guaranteed time window and uploading to a central cloud platform comprises: determining a priority queue for the alarm message based on the dynamic risk level and time sequence; based on the priority queue, sequentially embedding the alarm message in the priority queue into a guaranteed time window from the head to the tail to obtain early warning information; uploading the early warning information to the central cloud platform.

[0014] In addition, to achieve the above-mentioned purpose, the present application also proposes a safety early warning device for a smart rail area based on edge computing, which comprises: a data acquisition module, configured to acquire video stream, three-dimensional vibration spectrum and micro-environment parameter set in a covered area in real time, and construct a multi-modal data cube with spatio-temporal correlation according to the video stream, the three-dimensional vibration spectrum and the micro-environment parameter set; a feature processing module, configured to extract deep features of each modality in the multi-modal data cube, dynamically weight the deep features, and generate a risk feature tensor with spatio-temporal consistency; a risk rating module, configured to determine a neighborhood risk factor based on the risk feature tensor, and determine a risk positioning coordinate and a dynamic risk level based on the neighborhood risk factor; an information encryption module, configured to post-quantum encrypt the risk positioning coordinate and the dynamic risk level to obtain a risk fingerprint, and generate an alarm message based on the risk fingerprint; a safety early warning module, configured to embed the alarm message into a guaranteed time window and upload to a central cloud platform, so that the central cloud platform triggers an early warning response.

[0015] In addition, to achieve the above-mentioned purpose, the present application also proposes a safety early warning device for a smart rail area based on edge computing, which comprises: a memory, a processor and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the safety early warning method for a smart rail area based on edge computing as described above.

[0016] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer readable storage medium, and the storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the safety early warning method for a smart rail area based on edge computing as described above.

[0017] In addition, to achieve the above-mentioned purpose, the application further provides a computer program product, which comprises a computer program, and the computer program realizes the steps of the edge computing-based intelligent rail area safety early warning method as above when executed by a processor.

[0018] The application provides an edge computing-based intelligent rail area safety early warning method, which realizes real-time collection of video streams, three-dimensional vibration spectra and micro-environment parameter sets in a covered area, construction of a multi-modal data cube with space-time correlation according to the video streams, three-dimensional vibration spectra and micro-environment parameter sets, extraction of deep features of each mode in the multi-modal data cube, dynamic weighting of the deep features, generation of a risk feature tensor with space-time consistency, determination of a neighborhood risk factor based on the risk feature tensor, determination of a risk positioning coordinate and a dynamic risk level based on the neighborhood risk factor, post-quantum encryption of the risk positioning coordinate and the dynamic risk level, obtaining of a risk fingerprint, generation of an alarm message based on the risk fingerprint, embedding of the alarm message into a guarantee time window and uploading of the alarm message to a central cloud platform to enable the central cloud platform to trigger an early warning response. The application realizes accurate, real-time and safe perception and response to risks. BRIEF DESCRIPTION OF DRAWINGS

[0019] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and serve to explain the principles of the application together with the specification.

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, for those skilled in the art, other drawings can also be obtained based on these drawings without any creative effort.

[0021] Figure 1 Flowchart of the edge computing-based intelligent rail area safety early warning method of the first embodiment of the application; Figure 2 Data processing flowchart of the edge computing-based intelligent rail area safety early warning method of the first embodiment of the application; Figure 3 Module structure diagram of the edge computing-based intelligent rail area safety early warning device of the embodiment of the application; Figure 4 Device structure diagram of the hardware running environment involved in the edge computing-based intelligent rail area safety early warning method of the embodiment of the application.

[0022] The implementation of the purpose of the application, functional features and advantages will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0023] It should be understood that the specific embodiments described herein are merely intended to explain the technical solutions of the present application, and are not intended to limit the present application.

[0024] In order to better understand the technical solutions of the present application, the following will be described in detail in combination with the drawings of the specification and specific embodiments.

[0025] The main solution of the embodiments of the present application is: real-time collection of video stream, three-dimensional vibration spectrum and micro-environment parameter set in the covered area, construction of a multi-modal data cube with spatio-temporal correlation according to the video stream, the three-dimensional vibration spectrum and the micro-environment parameter set; Extraction of deep features of each modality in the multi-modal data cube, dynamic weighting of the deep features, generation of a risk feature tensor with spatio-temporal consistency; Determination of a neighborhood risk factor based on the risk feature tensor, determination of a risk positioning coordinate and a dynamic risk level based on the neighborhood risk factor; Post-quantum encryption of the risk positioning coordinate and the dynamic risk level to obtain a risk fingerprint, generation of an alarm message based on the risk fingerprint; Embedding the alarm message into a guarantee time window and uploading it to a central cloud platform to make the central cloud platform trigger an early warning response.

[0026] At present, with the development of rail transit towards high speed and high density, the safe operation and maintenance of the rail area (including tracks, tunnels, stations, etc.) is facing unprecedented challenges. The existing system mostly uses isolated monitoring means, and false positives or false negatives often occur, so the existing rail area safety warning technology cannot meet the safety needs.

[0027] The present application provides a solution by real-time collection of video stream, three-dimensional vibration spectrum and micro-environment parameter set in the covered area, construction of a multi-modal data cube with spatio-temporal correlation according to the video stream, the three-dimensional vibration spectrum and the micro-environment parameter set, extraction of deep features of each modality in the multi-modal data cube, dynamic weighting of the deep features, generation of a risk feature tensor with spatio-temporal consistency, determination of a neighborhood risk factor based on the risk feature tensor, determination of a risk positioning coordinate and a dynamic risk level based on the neighborhood risk factor, post-quantum encryption of the risk positioning coordinate and the dynamic risk level to obtain a risk fingerprint, generation of an alarm message based on the risk fingerprint, embedding the alarm message into a guarantee time window and uploading it to a central cloud platform to make the central cloud platform trigger an early warning response. The precise, real-time and safe perception and response to risks are realized.

[0028] It should be noted that the execution subject of the embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, or an electronic device capable of realizing the above functions, an edge computing-based intelligent rail area safety warning device, etc. The embodiment does not make specific limitations on this. The following takes the edge computing-based intelligent rail area safety warning device as an example to describe the embodiment and the following embodiments.

[0029] All actions of obtaining signals, information or data in this application are carried out in accordance with the corresponding data protection regulations and policies of the country where the device is located, and with the authorization given by the corresponding device owner.

[0030] The embodiment of the present application provides an edge computing-based intelligent rail area safety warning method, which refers to Figure 1 , Figure 1 The flowchart of the first embodiment of the edge computing-based intelligent rail area safety warning method of the present application is shown in the figure.

[0031] In the embodiment, the edge computing-based intelligent rail area safety warning method includes steps S10-S50: Step S10, real-time collection of video stream, three-dimensional vibration spectrum and micro-environment parameter set in the covered area, construction of spatio-temporal correlation multi-modal data cube according to the video stream, the three-dimensional vibration spectrum and the micro-environment parameter set; It should be noted that the physical area in the intelligent rail area where sensors, cameras and other monitoring devices are deployed usually includes tracks, tunnels, platforms, switches and other key sections. The area that can be monitored by the monitoring devices is the covered area. The video stream, the three-dimensional vibration spectrum and the micro-environment parameter set are different forms of data collected by different monitoring devices. The video stream refers to a continuous image sequence collected in real time by a high-definition camera, which is used to monitor visual information such as personnel activities, equipment status and foreign object intrusion in the rail area. The three-dimensional vibration spectrum refers to the frequency spectrum features obtained by Fourier transform of the vibration signals collected by the vibration sensor in three orthogonal directions, which is used to analyze the type, intensity and position of the vibration source. The micro-environment parameter set includes temperature, humidity, air pressure, light intensity, wind speed and other environmental monitoring data, which is used to evaluate the environmental status of the rail area and its impact on equipment safety. The spatio-temporal correlation multi-modal data cube is a data structure that aligns and fuses data from different sources and different modalities in time and space dimensions.

[0032] It can be understood that the video stream is captured in real time by high-definition cameras deployed in the track area, and is initially compressed and denoised by the edge node; the three-dimensional vibration spectrum is collected by the vibration sensor, and the signal is filtered and Fourier transformed by the edge computing node to extract the frequency domain features; the microenvironmental parameters are collected by various environmental sensors and calibrated and normalized by the edge node. After the data is collected and related preprocessing is completed, it can be fused according to the multimodal data to construct a multimodal data cube. At this time, time synchronization, spatial alignment and data fusion are required to obtain a data cube. The data cube can be defined as: .

[0033] in, is a data cube, is the time dimension, which represents the timestamp sequence of data collection. is the spatial dimension, representing the three-dimensional coordinate in the track area, is the modal dimension, representing different data types, Represents the length of each dimension respectively.

[0034] In a feasible embodiment, the steps of collecting the video stream, the three-dimensional vibration spectrum and the microenvironment parameter set in the coverage area in real time and constructing a spatiotemporally correlated multimodal data cube based on the video stream, the three-dimensional vibration spectrum and the microenvironment parameter set include: Aligning the clock information of all sensors in the coverage area, establishing a global timestamp based on the clock information, mapping the location information of all sensors to a three-dimensional spatial coordinate system, and generating a unified spatiotemporal network based on the global timestamp and the three-dimensional spatial coordinate system; Perform dynamic background modeling and foreground segmentation on the video stream of the coverage area, determine the moving targets in the area, and determine the spatiotemporal dynamic characteristics of the moving targets; Performing wavelet packet decomposition on the three-dimensional vibration spectrum of the coverage area to extract impact energy features, and constructing an abnormal pattern recognition matrix based on the impact energy features; The microenvironmental parameter set is fused through the Kalman filter to generate an environmental risk index; In the unified spatiotemporal network, a spatiotemporal-associated multimodal data cube is constructed according to the spatiotemporal dynamic characteristics, the abnormal pattern recognition matrix, and the environmental risk index.

[0035] It should be noted that the clock information refers to the local time information recorded by the internal clock of each sensor, and there is a problem of time asynchronization due to hardware differences or network delay. The global timestamp is a unified time reference system generated by aligning the clocks of all sensors, which can ensure that all data are consistent in the time dimension. The three-dimensional spatial coordinate system is a coordinate system established according to the actual geographical space of the track area, which is used to uniformly represent the physical positions of sensors and targets. The space-time unified network is a structured data network that associates time and space, which is used to integrate multi-source sensor data.

[0036] It can be understood that the space-time dynamic feature is the behavior feature of the moving target extracted from the video in time and space, such as motion trajectory, speed, acceleration, direction, etc. The abnormal pattern recognition matrix is a matrix constructed by feature extraction on the vibration signal, which is used to identify abnormal vibration patterns such as impact, fracture and collision. The environmental risk index is a quantitative index obtained by fusion calculation on micro-environmental parameters, which is used to evaluate the influence degree of environment on track safety.

[0037] In specific implementation, refer to Figure 2 , Figure 2 for a data processing flow diagram. When time synchronization is performed, NTP (Network Time Protocol) or PTP (Precision Time Protocol) is used to align the time of all sensors, and a global timestamp t g is generated. Then the positions of all sensors are mapped to a three-dimensional spatial coordinate system, which can be mapped to a unified three-dimensional coordinate system through laser scanning or visual calibration, and the position of each sensor is denoted as . Then a space-time unified network is generated according to the global timestamp and the three-dimensional spatial coordinate system, and the space-time unified network can be represented as a graph structure . Wherein the node is a sensor or a spatial grid, and the edge represents the space-time association relationship. Then, dynamic background modeling and foreground segmentation are performed on the video stream in the covered area. When dynamic background modeling is performed, Gaussian Mixture Model (GMM) can be used to model the video background, and the model can be represented as:

[0038] Wherein, I t is the current video frame, K is the number of Gaussian components, N(·) is the Gaussian distribution, ω k,t is the weight, μ k,t is the mean, and Σ k,t is the covariance.

[0039] When performing foreground segmentation, a moving target can be extracted by background subtraction to generate a binary mask , determine the moving target in the region. Then determine the spatio-temporal dynamic characteristics of each moving target, calculate the trajectory of the moving target , and extract the speed , acceleration .

[0040] When wavelet packet decomposition is performed on the three-dimensional vibration spectrum of the coverage area, the three-dimensional vibration signal s(t)=[s x (t),s y (t),s z (t)] is subjected to wavelet packet transform to extract the frequency band energy feature E j,k :

[0041] Where W j,k is the wavelet transform coefficient, j is the decomposition level, and k is the frequency band index.

[0042] The features of the high frequency band are extracted from the frequency band energy features to obtain the impact energy features. Then an abnormal pattern recognition matrix A is constructed according to the impact energy features. Then the microenvironment parameter set is interpolated and compensated to obtain a microenvironment parameter preprocessing set, a state vector is determined according to the microenvironment parameter preprocessing set, the state vector is fused by the observation equation of the Kalman filter to obtain a microenvironment filtering parameter, and an environmental risk index is generated according to the environmental risk weight factor and the microenvironment filtering parameter.

[0043] The microenvironment parameter set is fused by the Kalman filter to estimate the state of the temperature T, humidity H and wind speed W in the microenvironment parameter set, and the filtered environmental parameters .

[0044]

[0045]

[0046] Where, is the state vector, is the observation vector, and are state transition and observation matrices, and are noises.

[0047] The environmental risk index R env is calculated after fusing the environmental parameters, and the calculation formula of the environmental risk index is: .

[0048] Where, is the weight coefficient.

[0049] According to the spatio-temporal dynamic characteristics, the abnormal pattern recognition matrix and the environmental risk index, a spatio-temporal correlation multi-modal data cube is constructed.

[0050] In a feasible implementation, the step of dynamically modeling the background and segmenting the foreground of the video stream of the coverage area, determining the moving target in the area, and determining the spatio-temporal dynamic characteristics of the moving target comprises: determining the light intensity in the video stream of the coverage area, generating a dynamic light intensity weight according to the light intensity, and creating a dynamic background based on the dynamic light intensity weight; performing a morphological top-hat transformation on the video stream, determining the contrast of small targets in the video stream, determining a foreground mask according to the contrast, performing foreground segmentation based on the foreground mask, and determining a foreground target; projecting the foreground target to the dynamic background, and determining the moving target in the area; calculating the minimum circumscribed rectangle of the moving target based on the foreground mask, extracting a feature vector of the minimum circumscribed rectangle, and generating spatio-temporal dynamic characteristics according to the feature vector.

[0051] It should be noted that the dynamic light intensity weight is a weight coefficient calculated according to the real-time light intensity in the video stream, which is used to adjust the adaptability of the background model to light changes and ensure the stability of the background modeling under different light conditions. The dynamic background is a background model that can adaptively update over time, which effectively deals with scene changes such as light changes and shadow interference by fusing historical frame information and current light conditions. The foreground mask is a binary image, where the pixel points with a value of 1 represent the moving target, and the pixel points with a value of 0 represent the background. The foreground target is a moving object segmented from the video stream, such as personnel, vehicles, foreign objects, etc., which is the object that needs to be monitored and analyzed.

[0052] In a specific implementation, the current video frame The average gray value or luminance component is calculated to obtain the light intensity The dynamic light intensity weight is generated according to the light intensity :

[0053] wherein, is the moving average value of the recent light intensity, is an adjustment parameter.

[0054] Then a dynamic background is created according to the dynamic light intensity weight The equation is constructed as: .

[0055] And the video stream is subjected to morphological top-hat transformation, the contrast of small target and background is enhanced, the contrast of small target in the video stream is determined, the foreground mask is determined according to the contrast, and the mask generation formula is:

[0056] Wherein, T contrast is a contrast threshold.

[0057] The mask is subjected to connected region analysis, and the continuous foreground region is extracted as a candidate moving target. The foreground target is projected to the dynamic background , and the coincidence degree is calculated. If the coincidence degree is low, it is confirmed as a real moving target. The minimum circumscribed rectangle of each moving target is calculated to obtain parameters such as position, width, height, and angle. Then, the feature vector F is extracted: .

[0058] Wherein, is the target center coordinate, is the rectangular width and height, is the rectangular rotation angle, is the target speed in the x and y directions, is the target acceleration in the x and y directions.

[0059] In a feasible implementation, the step of performing wavelet packet decomposition on the three-dimensional vibration spectrum of the coverage area, extracting an impact energy feature, and constructing an anomaly pattern recognition matrix based on the impact energy feature includes: Performing adaptive noise complete ensemble empirical mode decomposition on the three-dimensional vibration spectrum of the coverage area to generate an enhanced signal; Performing wavelet packet decomposition on the enhanced signal to obtain a plurality of sub-bands; Calculating the energy entropy of each of the sub-bands, and marking the sub-band of a preset frequency in the energy entropy as an impact sensitive frequency band; Determining the energy mutation rate of the impact sensitive frequency band, marking the impact sensitive frequency band as an anomaly pattern when the energy mutation rate is greater than an abnormal mutation rate, and extracting an impact energy feature of the impact sensitive frequency band; Determining the time feature, frequency domain feature, and spatial feature of the impact energy feature, and generating an anomaly pattern recognition matrix according to the time feature, the frequency domain feature, and the spatial feature.

[0060] It should be noted that the enhanced signal refers to the vibration signal processed by adaptive noise complete set empirical mode decomposition, which effectively separates the noise components in the original vibration data. Wavelet packet decomposition divides the signal into multiple frequency bands at multiple scales, obtaining multiple sub-bands, each of which corresponds to the component of the signal in a specific frequency range. The impact sensitive frequency band is a frequency band containing the main impact energy and high energy entropy, which is the key frequency band for impact feature extraction. The energy mutation rate is the impact sensitive frequency band whose energy mutation rate exceeds the threshold. The impact energy feature refers to the feature quantity extracted from the impact sensitive frequency band which can represent the impact event.

[0061] In a specific implementation, the adaptive noise complete set empirical mode decomposition is used to enhance the three-dimensional vibration spectrum signal:

[0062] wherein, is the original vibration signal, IMF i is the i-th intrinsic mode function, r n is the residual component.

[0063] The enhanced signal is reconstructed from the first k main IMF components:

[0064] wherein, x enhanced is the enhanced signal.

[0065] Then, the enhanced signal is decomposed by L-layer wavelet packet to obtain sub-bands: .

[0066] wherein, is the wavelet packet basis function, is the decomposition scale (j=1, 2,..., L), and k is the frequency band index (k=0, 1,..., -1).

[0067] The energy entropy of each sub-band is calculated:

[0068]

[0069]

[0070] wherein, E k is the energy of the k-th sub-band, P k is the energy proportion of the k-th sub-band, and H k is the energy entropy of the k-th sub-band.

[0071] The energy entropy H k ​Greater than the threshold H th The sub-frequency bands whose center frequencies are within the preset shock frequency range are marked as shock-sensitive frequency bands.

[0072] Calculate the energy mutation rate of the shock-sensitive frequency band :

[0073] exist When , the frequency band is marked as abnormal mode, is the abnormal mutation rate threshold, determined based on historical data and experiments. Then, three types of impact energy features are extracted: time feature , frequency domain characteristics and spatial characteristics . And construct the abnormal pattern recognition matrix B:

[0074] Step S20: extracting deep features of each modality in the multimodal data cube, dynamically weighting the deep features, and generating a risk feature tensor with spatiotemporal consistency; It should be noted that deep features refer to high-level, abstract feature representations extracted from raw multimodal data using deep learning networks. The risk feature tensor is a multidimensional data structure generated by dynamically weighted fusion of the deep features of each modality. It is used to comprehensively represent the safety risk status of the track area.

[0075] It can be understood that the deep features of each modality in the multimodal data cube are extracted. For the visual feature branch, a 3D CNN is used to extract spatiotemporal features. For the vibration feature branch, a time-series CNN + LSTM network is used to process vibration data. A fully connected deep network is used to process the environmental feature branch. Then, dynamic calculations are performed using the attention mechanism to generate a risk feature tensor with spatiotemporal consistency.

[0076] In a feasible implementation, the steps of extracting deep features of each modality in the multimodal data cube, dynamically weighting the deep features, and generating a risk feature tensor with spatiotemporal consistency include: Extracting spatial features of the video stream and spatiotemporal features of the vibration spectrum in the multimodal data cube through dual-path convolution; Performing feature alignment on the spatial features and the spatiotemporal features to obtain spatiotemporally aligned depth features; determining an environmental risk index, and generating an attention weight based on the environmental risk index; The deep features are dynamically weighted based on the attention weights to generate a risk feature tensor with spatiotemporal consistency.

[0077] It should be noted that the spatial feature of the video stream refers to a feature extracted from the video data representing the distribution of visual content in two-dimensional space, mainly describing the shape, texture, edge and other static appearance attributes of the target. The spatio-temporal feature of the vibration spectrum refers to a feature that simultaneously captures the time evolution law and frequency distribution characteristics from the vibration signal, which contains both the dynamic change information of the vibration over time and the distribution characteristics in the frequency domain. The attention weight is used to measure the importance and reliability of different modal features under the current environmental conditions.

[0078] In a specific implementation, the spatial feature of the video stream and the spatio-temporal feature of the vibration spectrum in the multi-modal data cube are extracted by a double-path convolution. The video stream spatial feature extraction path uses a 2D CNN to extract spatial features:

[0079] wherein, represents a two-dimensional convolution operation, , are the weights and biases of the spatial convolution, is the video data in the data cube.

[0080] The vibration spectrum spatio-temporal feature extraction path uses a 3D CNN to extract spatio-temporal features: .

[0081] wherein, represents a three-dimensional convolution operation, is the vibration data in the data cube, and are the weights and biases of the spatio-temporal convolution.

[0082] Then the spatial features and the spatio-temporal features are aligned to obtain spatio-temporally aligned deep features.

[0083] At the same time, the environmental risk index is determined, and the attention weight is calculated based on the environmental risk index: .

[0084] wherein, is the attention weight, is an activation function, , are learnable parameters.

[0085] After obtaining the attention weight, the deep features can be dynamically weighted with the attention weight to generate a risk feature tensor with spatio-temporal consistency.

[0086] Step S30, determining a neighborhood risk factor based on the risk feature tensor, determining a risk positioning coordinate and a dynamic risk level based on the neighborhood risk factor; It should be noted that the neighborhood risk factor refers to a risk propagation index calculated based on the statistical quantity of the features in the neighborhood around the target position in the risk feature tensor, and is used to quantify the diffusion degree and influence range of the risk in space. The accurate position coordinates of the risk source are determined by analyzing the spatial distribution of the neighborhood risk factor.

[0087] It can be understood that the risk feature tensor is used to calculate the risk intensity index and the risk diffusion degree. The neighborhood risk factor of each spatial position (x, y, z) is calculated. It includes the local risk density, the risk gradient feature, and the neighborhood risk factor. The calculation formula of the local risk density is:

[0088] The calculation formula of the risk gradient feature is:

[0089] The calculation formula of the neighborhood risk factor is:

[0090] wherein, is the neighborhood radius, is the number of points in the neighborhood, is an adjustment parameter.

[0091] Then the position of the risk source is determined by finding the extreme value point of the neighborhood risk factor. The final positioning coordinates need to be determined through maximum value detection and sub-pixel accurate positioning. The specific implementation process is: The calculation formula of the maximum value detection is:

[0092] The calculation formula of the sub-pixel accurate positioning is:

[0093] The calculation formula of the final positioning coordinates is: .

[0094] In the dynamic risk level assessment, the risk level is divided based on the statistical characteristics of the neighborhood risk factor. First, the risk intensity index and the risk diffusion degree can be calculated and determined, and the dynamic risk level is determined according to the risk intensity index and the risk diffusion degree. The specific calculation formula is: The calculation formula of the risk intensity index is:

[0095] The calculation formula of the risk diffusion degree is: ​​​

[0096] Dynamic risk level The calculation formula is:

[0097] wherein, is a weight coefficient.

[0098] In a feasible implementation, the step of determining the neighborhood risk factor based on the risk feature tensor, determining the risk positioning coordinate and the dynamic risk level based on the neighborhood risk factor comprises: determining the physical connection relationship of each device in the coverage area, and determining the device topology relationship based on the physical connection relationship; determining the neighborhood relationship of each device based on the device topology relationship; spatiotemporal convolution is performed on the neighborhood relationship, and the neighborhood risk information of the target hop number is aggregated; determining the neighborhood risk factor and the risk gradient according to the neighborhood risk information; determining the risk positioning coordinate according to the risk gradient, and determining the dynamic risk level according to the neighborhood risk factor and the risk positioning coordinate.

[0099] It should be noted that the device topology relationship refers to the network structure formed by the physical connection relationship and the spatial position relationship between each monitoring device (sensor, camera, etc.) in the rail area, which is usually represented by a graph structure, wherein the node represents the device, and the edge represents the connection relationship between the devices. The neighborhood relationship is a set of adjacent devices of each device within a certain hop number range determined based on the device topology relationship. The neighborhood risk information is a comprehensive risk indicator obtained by aggregating the risk features of the target device and its neighborhood devices. The risk factor is a numerical indicator quantifying the risk degree of the device or the area. The risk gradient is a vector field representing the rate and direction of change of risk in space.

[0100] In a specific implementation, first, a device topology network is constructed according to the physical connection relationship of each sensor in the coverage area, the spatial correlation between devices is determined, and then the neighborhood range of each device is defined based on the topology relationship. The risk feature information of the neighborhood devices within a certain hop number is aggregated through a spatiotemporal convolution operation to form neighborhood risk data reflecting the risk propagation characteristics. The neighborhood risk factor representing the local risk intensity and the risk gradient vector indicating the risk change direction are calculated according to these aggregated data. Finally, the accurate spatial coordinates of the risk source are determined by reverse tracking using the risk gradient field, and the risk source positioning result is comprehensively considered with the neighborhood risk factor value to dynamically generate a level indicator quantifying the risk severity.

[0101] In step S40, the risk positioning coordinate and the dynamic risk level are post-quantum encrypted to obtain a risk fingerprint, and an alarm message is generated based on the risk fingerprint.

[0102] It should be noted that the risk fingerprint refers to a unique digital identifier generated by performing post-quantum encryption and hash operation on the risk positioning coordinates and the dynamic risk level. The alarm message contains a structured data packet of risk fingerprint and encrypted risk information, which is used to transmit risk warning information in the system. The encrypted information refers to the ciphertext data obtained by encrypting the original risk data using a post-quantum cryptographic algorithm.

[0103] It can be understood that the specific process of generating an alarm message based on the risk fingerprint can be described as follows: performing post-quantum encryption on the risk positioning coordinates and the dynamic risk level to obtain encrypted information; performing hash operation on the encrypted information to obtain the risk fingerprint; and encapsulating the risk fingerprint and the encrypted information to obtain the alarm information.

[0104] In a specific implementation, the risk positioning coordinates and the dynamic risk level are encrypted using a post-quantum encryption algorithm to generate ciphertext data with quantum computing attack resistance, and then a digital fingerprint of the ciphertext data is extracted through hash operation to form a unique risk identifier. Finally, the encrypted data and the risk fingerprint are encapsulated with metadata such as a timestamp to generate a standardized alarm message, ensuring the confidentiality, integrity, and verifiability of the risk information during transmission.

[0105] Step S50, embedding the alarm message in a protection time window and uploading it to a central cloud platform, so that the central cloud platform triggers a warning response.

[0106] It should be noted that the protection time window refers to the maximum delay threshold allowed for data transmission. The central cloud platform refers to a remote server cluster with large-scale data storage and processing capabilities.

[0107] It can be understood that first, the generated alarm message is bound with a preset protection time window to ensure that the alarm information is transmitted within a specified time limit; then the message with the bound time window is uploaded to the central cloud platform through a secure communication protocol; after receiving the message, the central cloud platform verifies the validity of the time window, parses the encrypted risk information, and finally automatically triggers the corresponding warning response mechanism according to the parsing result, realizing the cloud collaborative disposal of risk events.

[0108] In one possible implementation, the step of embedding the alarm message in a protection time window and uploading it to a central cloud platform includes: determining a priority queue based on the dynamic risk level and the time sequence; based on the priority queue, sequentially embedding the alarm messages in the priority queue from the head to the tail of the queue into a protection time window to obtain warning information; The pre-warning information is uploaded to a central cloud platform.

[0109] In a specific implementation, first, a plurality of alarms are sorted according to dynamic risk levels contained in alarm messages and time sequences in which the dynamic risk levels are generated, to form a priority queue; then, a corresponding guarantee time window is embedded into each alarm message in sequence according to the queue, to form pre-warning information with time effectiveness constraints; finally, the packaged pre-warning information is uploaded to a central cloud platform through a network, and the platform receives and analyzes the information within the time window and triggers a corresponding pre-warning response mechanism.

[0110] The embodiment provides a smart rail area safety pre-warning method based on edge computing, real-time video stream, three-dimensional vibration spectrum and micro-environment parameter set in a covered area are collected, a multi-modal data cube with space-time correlation is constructed according to the video stream, the three-dimensional vibration spectrum and the micro-environment parameter set, deep features of each mode in the multi-modal data cube are extracted, the deep features are dynamically weighted, a risk feature tensor with space-time consistency is generated, a neighborhood risk factor is determined based on the risk feature tensor, a risk positioning coordinate and a dynamic risk level are determined based on the neighborhood risk factor, the risk positioning coordinate and the dynamic risk level are post-quantum encrypted, a risk fingerprint is obtained, an alarm message is generated based on the risk fingerprint, the alarm message is embedded into a guarantee time window and uploaded to a central cloud platform, so that the central cloud platform triggers a pre-warning response. Precise, real-time and safe perception and response to risks are realized.

[0111] It should be noted that the above examples are only used for understanding the present application and do not constitute a limitation on the smart rail area safety pre-warning method based on edge computing of the present application, and more forms of simple transformation based on the technical concept are within the protection scope of the present application.

[0112] The present application also provides a smart rail area safety pre-warning device based on edge computing, please refer to Figure 3 The smart rail area safety pre-warning device based on edge computing comprises: A data collection module 10 is configured to collect real-time video stream, three-dimensional vibration spectrum and micro-environment parameter set in a covered area, and construct a multi-modal data cube with space-time correlation according to the video stream, the three-dimensional vibration spectrum and the micro-environment parameter set; A feature processing module 20 is configured to extract deep features of each mode in the multi-modal data cube, dynamically weight the deep features, and generate a risk feature tensor with space-time consistency; A risk rating module 30 is configured to determine a neighborhood risk factor based on the risk feature tensor, and determine a risk positioning coordinate and a dynamic risk level based on the neighborhood risk factor; The information encryption module 40 is configured to perform post-quantum encryption on the risk positioning coordinates and the dynamic risk level to obtain a risk fingerprint, and generate an alarm message based on the risk fingerprint; The security warning module 50 is configured to embed the alarm message into a guarantee time window and upload the alarm message to a central cloud platform, so that the central cloud platform triggers a warning response.

[0113] In an embodiment, the data acquisition module 10 is further configured to align clock information of all sensors in a coverage area, establish a global timestamp based on the clock information, map position information of all sensors to a three-dimensional spatial coordinate system, and generate a space-time unified network based on the global timestamp and the three-dimensional spatial coordinate system; The video stream of the coverage area is subjected to dynamic background modeling and foreground segmentation, a moving target in the area is determined, and a space-time dynamic feature of the moving target is determined; The three-dimensional vibration spectrum of the coverage area is subjected to wavelet packet decomposition, an impact energy feature is extracted, and an abnormal pattern recognition matrix is constructed based on the impact energy feature; The micro-environment parameter set is fused by a Kalman filter to generate an environmental risk index; In the space-time unified network, a space-time correlated multi-modal data cube is constructed based on the space-time dynamic feature, the abnormal pattern recognition matrix, and the environmental risk index.

[0114] In an embodiment, the data acquisition module 10 is further configured to determine an illumination intensity in the video stream of the coverage area, generate a dynamic light intensity weight based on the illumination intensity, and create a dynamic background based on the dynamic light intensity weight; The video stream is subjected to morphological top-hat transformation, a contrast of small targets in the video stream is determined, a foreground mask is determined based on the contrast, foreground segmentation is performed based on the foreground mask, and a foreground target is determined; The foreground target is projected to the dynamic background to determine a moving target in the area; A minimum circumscribed rectangle of the moving target is calculated based on the foreground mask, a feature vector of the minimum circumscribed rectangle is extracted, and a space-time dynamic feature is generated based on the feature vector.

[0115] In an embodiment, the data acquisition module 10 is further configured to perform adaptive noise complete ensemble empirical mode decomposition on the three-dimensional vibration spectrum of the coverage area to generate an enhanced signal; The enhanced signal is subjected to wavelet packet decomposition to obtain a plurality of sub-bands; Energy entropy of each of the sub-bands is calculated, and a sub-band with a preset frequency in the energy entropy is marked as an impact-sensitive frequency band; determine an energy mutation rate of the impact-sensitive frequency band, mark the impact-sensitive frequency band as an abnormal pattern when the energy mutation rate is greater than an abnormal mutation rate, and extract an impact energy feature of the impact-sensitive frequency band; determine a time feature, a frequency domain feature, and a space feature of the impact energy feature, and generate an abnormal pattern recognition matrix according to the time feature, the frequency domain feature, and the space feature.

[0116] In a feasible implementation, the data acquisition module 10 is further configured to perform interpolation compensation on the micro-environment parameter set to obtain a micro-environment parameter pre-processing set. determine a state vector according to the micro-environment parameter pre-processing set, fuse the state vector by using an observation equation of the Kalman filter, and obtain a micro-environment filtering parameter; generate an environmental risk index according to the environmental risk weight factor and the micro-environment filtering parameter.

[0117] In a feasible implementation, the feature processing module 20 is further configured to extract a space feature of a video stream and a space-time feature of a vibration spectrum in the multi-modal data cube by using a double-path convolution. align the space feature and the space-time feature to obtain a space-time aligned deep feature; determine an environmental risk index, and generate an attention weight according to the environmental risk index; perform dynamic weighting on the deep feature based on the attention weight to generate a risk feature tensor with space-time consistency.

[0118] In a feasible implementation, the risk rating module 30 is further configured to determine a physical connection relationship of each device in the coverage area, and determine a device topology relationship based on the physical connection relationship. determine a neighborhood relationship of each device based on the device topology relationship; perform space-time convolution on the neighborhood relationship to aggregate neighborhood risk information of a target hop count; determine a neighborhood risk factor and a risk gradient according to the neighborhood risk information; determine a risk positioning coordinate according to the risk gradient, and determine a dynamic risk level according to the neighborhood risk factor and the risk positioning coordinate.

[0119] In a feasible implementation, the information encryption module 40 is further configured to perform post-quantum encryption on the risk positioning coordinate and the dynamic risk level to obtain encrypted information. perform a hash operation on the encrypted information to obtain a risk fingerprint; package the risk fingerprint and the encrypted information to obtain alarm information.

[0120] In a feasible implementation, the security early warning module 50 is further configured to determine a priority queue for the warning messages based on the dynamic risk level and the time sequence; Based on the priority queue, the warning messages in the priority queue are sequentially embedded into a guarantee time window from the head to the tail of the priority queue to obtain early warning information; The early warning information is uploaded to a central cloud platform.

[0121] The edge computing-based intelligent rail area security early warning device provided in the application adopts the edge computing-based intelligent rail area security early warning method in the above embodiment, and can solve the technical problem of low early warning efficiency caused by data isolation in rail area security monitoring. Compared with the prior art, the edge computing-based intelligent rail area security early warning device provided in the application has the same beneficial effects as the edge computing-based intelligent rail area security early warning method provided in the above embodiment, and other technical features in the edge computing-based intelligent rail area security early warning device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0122] The application provides an edge computing-based intelligent rail area security early warning device. The edge computing-based intelligent rail area security early warning device comprises at least one processor and a memory in communication connection with the at least one processor. The memory stores instructions executable by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to execute the edge computing-based intelligent rail area security early warning method in the above embodiment one.

[0123] Reference will be made to the following description Figure 4 which shows a structural schematic diagram of an edge computing-based intelligent rail area security early warning device suitable for implementing the embodiments of the application. The edge computing-based intelligent rail area security early warning device in the embodiments of the application can include but is not limited to mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (Personal Digital Assistant), PADs (Portable Application Description), PMPs (Portable Media Player), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), and the like, and fixed terminals such as digital TVs, desktop computers, and the like. Figure 4 The edge computing-based intelligent rail area security early warning device shown is only an example, and should not impose any limitation on the functions and use range of the embodiments of the application.

[0124] As Figure 4As shown, the edge computing based intelligent rail area safety warning device can include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to programs stored in a ROM (Read Only Memory) 1002 or programs loaded from a storage device 1003 into a RAM (Random Access Memory) 1004. In the RAM 1004, various programs and data required for operation of the edge computing based intelligent rail area safety warning device are also stored. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, an LCD (Liquid Crystal Display), a speaker, a vibrator, etc.; the storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the edge computing based intelligent rail area safety warning device to communicate wirelessly or wired with other devices to exchange data. Although the edge computing based intelligent rail area safety warning device with various systems is shown in the figure, it should be understood that all the shown systems are not required to be implemented or possessed. More or less systems can be alternatively implemented or possessed.

[0125] In particular, the processes described above with reference to the flowcharts can be implemented as a computer software program according to embodiments of the present disclosure. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network through the communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the methods of embodiments of the present disclosure are performed.

[0126] The edge computing-based intelligent rail area safety warning device provided in the application adopts the edge computing-based intelligent rail area safety warning method in the above embodiment, and can solve the technical problem of the edge computing-based intelligent rail area safety warning. Compared with the prior art, the edge computing-based intelligent rail area safety warning device provided in the application has the same beneficial effects as the edge computing-based intelligent rail area safety warning method provided in the above embodiment, and other technical features in the edge computing-based intelligent rail area safety warning device are the same as the features disclosed in the previous embodiment method, which will not be repeated here.

[0127] It should be understood that various parts of the present application can be realized by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0128] The above is merely specific implementation of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0129] The present application provides a computer readable storage medium having computer readable program instructions (i.e. computer programs) stored thereon, the computer readable program instructions being used to execute the edge computing-based intelligent rail area safety warning method in the above embodiment.

[0130] The computer readable storage medium provided in the application may be, for example, a U disk, but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination thereof. More specific examples of the computer readable storage medium may include, but are not limited to, an electrical connection with one or more conductive wires, a portable computer disk, a hard disk, a RAM (Random Access Memory), a ROM (Read Only Memory), an EPROM (Erasable Programmable Read Only Memory or flash memory), an optical fiber, a CD-ROM (CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the embodiment, the computer readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system, or device. The program code contained on the computer readable storage medium can be transmitted by any suitable medium, including but not limited to an electrical wire, an optical cable, an RF (Radio Frequency), and the like, or any suitable combination thereof.

[0131] The computer readable storage medium described above may be contained in the intelligent rail area safety warning device based on edge computing; or may exist independently without being assembled into the intelligent rail area safety warning device based on edge computing.

[0132] The computer readable storage medium described above carries one or more programs, when the one or more programs are executed by the intelligent rail area safety warning device based on edge computing, the intelligent rail area safety warning device based on edge computing: collects video streams, three-dimensional vibration spectra, and micro-environment parameter sets in a coverage area in real time, constructs a multi-modal data cube with space-time correlation according to the video streams, the three-dimensional vibration spectra, and the micro-environment parameter sets; extracts deep features of each modality in the multi-modal data cube, dynamically weights the deep features, and generates a risk feature tensor with space-time consistency; determines a neighborhood risk factor based on the risk feature tensor, determines a risk positioning coordinate and a dynamic risk level based on the neighborhood risk factor; performs post-quantum encryption on the risk positioning coordinate and the dynamic risk level to obtain a risk fingerprint, generates an alarm message based on the risk fingerprint; embeds the alarm message into a guarantee time window and uploads it to a central cloud platform, so that the central cloud platform triggers a warning response.

[0133] The computer program code for performing the operations of the present application may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a LAN (Local Area Network) or a WAN (Wide Area Network), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0134] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0135] The modules described in the embodiments of the present application may be implemented in software or hardware, wherein the name of a module does not necessarily limit the unit itself.

[0136] The computer-readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the aforementioned edge computing-based smart rail zone safety warning method. This computer-readable storage medium can address the technical issues surrounding edge computing-based smart rail zone safety warning. Compared to the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the edge computing-based smart rail zone safety warning method provided in the aforementioned embodiments, and are not further elaborated here.

[0137] The application also provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the edge computing-based safety early warning method for a smart rail area as described above.

[0138] The computer program product provided by the application can solve the technical problem of the edge computing-based safety early warning for a smart rail area. Compared with the prior art, the beneficial effects of the computer program product provided by the application are the same as those of the edge computing-based safety early warning method for a smart rail area provided by the above-mentioned embodiments, and are not described here.

[0139] The above is only some embodiments of the application, and does not limit the patent scope of the application. Any equivalent structural transformation made by using the content of the application specification and drawings, or direct / indirect application in other related technical fields is included in the patent protection scope of the application.

Claims

1. A method for safety warning in a smart rail area based on edge computing, characterized in that, The edge computing-based intelligent rail area safety warning method comprises: Real-time collection of video stream, three-dimensional vibration spectrum and micro-environment parameter set in the covered area, construction of a multi-modal data cube with spatiotemporal correlation according to the video stream, the three-dimensional vibration spectrum and the micro-environment parameter set; Extraction of deep features of each modality in the multi-modal data cube, dynamic weighting of the deep features, and generation of a risk feature tensor with spatiotemporal consistency; Determination of a neighborhood risk factor based on the risk feature tensor, determination of a risk positioning coordinate and a dynamic risk level based on the neighborhood risk factor; Post-quantum encryption of the risk positioning coordinate and the dynamic risk level to obtain a risk fingerprint, and generation of an alarm message based on the risk fingerprint; Embedding of the alarm message into a guarantee time window and uploading to a central cloud platform to enable the central cloud platform to trigger a warning response.

2. The method of claim 1, wherein, The step of real-time collection of video stream, three-dimensional vibration spectrum and micro-environment parameter set in the covered area, and construction of a multi-modal data cube with spatiotemporal correlation according to the video stream, the three-dimensional vibration spectrum and the micro-environment parameter set comprises: Alignment of clock information of all sensors in the covered area, establishment of a global timestamp according to the clock information, mapping of position information of all sensors to a three-dimensional space coordinate system, and generation of a spatiotemporal unified network according to the global timestamp and the three-dimensional space coordinate system; Dynamic background modeling and foreground segmentation of the video stream of the covered area, determination of a moving target in the area, and determination of spatiotemporal dynamic features of the moving target; Wavelet packet decomposition of the three-dimensional vibration spectrum of the covered area, extraction of impact energy features, and construction of an abnormal pattern recognition matrix based on the impact energy features; Fusion of the micro-environment parameter set by a Kalman filter to generate an environmental risk index; In the spatiotemporal unified network, construction of a multi-modal data cube with spatiotemporal correlation according to the spatiotemporal dynamic features, the abnormal pattern recognition matrix and the environmental risk index.

3. The method of claim 2, wherein, The step of dynamic background modeling and foreground segmentation of the video stream of the covered area, determination of a moving target in the area, and determination of spatiotemporal dynamic features of the moving target comprises: Determination of illumination intensity in the video stream of the covered area, generation of a dynamic light intensity weight according to the illumination intensity, and creation of a dynamic background based on the dynamic light intensity weight; Morphological top-hat transformation of the video stream, determination of contrast of small targets in the video stream, determination of a foreground mask according to the contrast, foreground segmentation based on the foreground mask, and determination of a foreground target; Projection of the foreground target to the dynamic background to determine a moving target in the area; Calculation of a minimum bounding rectangle of the moving target based on the foreground mask, extraction of a feature vector of the minimum bounding rectangle, and generation of spatiotemporal dynamic features according to the feature vector.

4. The method of claim 2, wherein, The step of wavelet packet decomposition of the three-dimensional vibration spectrum of the covered area, extraction of impact energy features, and construction of an abnormal pattern recognition matrix based on the impact energy features comprises: Adaptive noise complete ensemble empirical mode decomposition of the three-dimensional vibration spectrum of the covered area to generate an enhanced signal; Wavelet packet decomposition of the enhanced signal to obtain a plurality of sub-bands; Calculate energy entropy of each sub-band, mark the sub-band with preset frequency as impact sensitive frequency band in the energy entropy; Determine the energy mutation rate of the impact sensitive frequency band, when the energy mutation rate is greater than the abnormal mutation rate, mark the impact sensitive frequency band as abnormal mode, and extract the impact energy feature of the impact sensitive frequency band; Determine the time feature, frequency domain feature and space feature of the impact energy feature, and generate an abnormal mode recognition matrix according to the time feature, the frequency domain feature and the space feature.

5. The method of claim 2, wherein, The step of fusing the micro environment parameter set through the Kalman filter to generate the environment risk index comprises: Interpolation compensation is performed on the micro environment parameter set to obtain a micro environment parameter pre-processing set; Determine a state vector according to the micro environment parameter pre-processing set, and fuse the state vector through an observation equation of the Kalman filter to obtain a micro environment filtering parameter; Generate an environment risk index according to an environment risk weight factor and the micro environment filtering parameter.

6. The method of claim 1, wherein, The step of extracting the depth feature of each modality in the multi-modal data cube, dynamically weighting the depth feature, and generating a risk feature tensor with spatio-temporal consistency comprises: Extract the spatial feature of the video stream and the spatio-temporal feature of the vibration spectrum in the multi-modal data cube through double-path convolution; Perform feature alignment on the spatial feature and the spatio-temporal feature to obtain a spatio-temporally aligned depth feature; Determine an environment risk index, and generate an attention weight according to the environment risk index; Dynamically weight the depth feature based on the attention weight to generate a risk feature tensor with spatio-temporal consistency.

7. The method of claim 1, wherein, The step of determining a neighborhood risk factor based on the risk feature tensor, determining a risk positioning coordinate and a dynamic risk level based on the neighborhood risk factor comprises: Determine the physical connection relationship of each device in the coverage area, and determine the device topology relationship based on the physical connection relationship; Determine the neighborhood relationship of each device based on the device topology relationship; Perform spatio-temporal convolution on the neighborhood relationship to aggregate the neighborhood risk information of the target hop number; Determine a neighborhood risk factor and a risk gradient according to the neighborhood risk information; Determine a risk positioning coordinate according to the risk gradient, and determine a dynamic risk level according to the neighborhood risk factor and the risk positioning coordinate.

8. The method of claim 1, wherein, The step of post-quantum encrypting the risk positioning coordinate and the dynamic risk level to obtain a risk fingerprint, and generating an alarm message based on the risk fingerprint comprises: Post-quantum encrypt the risk positioning coordinate and the dynamic risk level to obtain encrypted information; Perform a hash operation on the encrypted information to obtain a risk fingerprint; Package the risk fingerprint and the encrypted information to obtain an alarm message.

9. The method of claim 1, wherein, The step of embedding the alarm message in a protection time window and uploading it to a central cloud platform comprises: Determine a priority queue based on the dynamic risk level and time sequence of the alarm message; Based on the priority queue, sequentially embed the alarm messages in the priority queue from the head to the tail into a protection time window to obtain early warning information; Upload the early warning information to the central cloud platform.

10. An edge computing-based intelligent rail area safety warning device, characterized in that, The edge-computing-based intelligent rail area safety warning device comprises: a data acquisition module, configured to collect video stream, three-dimensional vibration spectrum and micro-environment parameter set in a covered area in real time, and construct a multi-modal data cube with space-time correlation according to the video stream, the three-dimensional vibration spectrum and the micro-environment parameter set; a feature processing module, configured to extract deep features of each mode in the multi-modal data cube, dynamically weight the deep features, and generate a risk feature tensor with space-time consistency; a risk rating module, configured to determine a neighborhood risk factor based on the risk feature tensor, determine a risk positioning coordinate and a dynamic risk level based on the neighborhood risk factor; an information encryption module, configured to post-quantum encrypt the risk positioning coordinate and the dynamic risk level to obtain a risk fingerprint, and generate an alarm message based on the risk fingerprint; a safety warning module, configured to upload the alarm message to a central cloud platform after embedding the alarm message in a guarantee time window, so that the central cloud platform triggers a pre-warning response.

Images