Road maintenance construction intelligent security management early warning method and system

Abstract

The invention relates to the technical field of traffic control, and discloses a road maintenance construction intelligent security management early warning method and system, and the method comprises the steps: obtaining historical state data and real-time monitoring data of a multi-section maintenance region, extracting operation duration features, personnel density features and equipment frequency features, and generating state vectors; further performing connected piece screening processing and mapping accumulation processing on the image connectivity data and the reflection amplitude data to obtain connected region features and reflection amplitude features, and performing multi-scale alignment processing on the state vector and the connected region features on the basis to obtain a state vector and reflection amplitude features; and performing segmented matching processing on the alignment feature sequence and the reflection amplitude feature to generate a matching matrix, thereby realizing dynamic fusion matching of the early warning level and the operation state. The interference value sequence is obtained by performing difference calculation processing on the matching matrix based on the preset adjacent matrix, and grade division is performed according to the interference value sequence, so that the disturbance change among the sections can be accurately identified.

Description

Technical Field

[0001] The present invention relates to the field of traffic control technology, and more specifically, to a method and system for intelligent security management and early warning of highway maintenance and construction. Background Art

[0002] During road construction or maintenance operations, in order to ensure the safety of construction workers and reduce the risk of traffic accidents, it is usually necessary to deploy a construction area early warning system to remind and guide passing vehicles. In the existing technology, construction early warning mostly relies on fixed alarm devices or vehicle-mounted signal equipment, such as flashing lights, sound and light alarms, electronic display boards, etc., and is supplemented by cameras or radars for traffic flow monitoring. Some advanced systems have introduced geographic fence technology or vehicle-road collaborative communication technology to achieve dynamic identification and status updates of designated construction areas, thereby improving the efficiency of early warning response. However, most existing systems are still based on single-point early warning and lack the ability to perceive and coordinate the collaborative status of multi-section construction.

[0003] However, in the scenario where maintenance operations are carried out on multiple sections of roads at the same time, the existing early warning system often has problems such as frequent false triggering, unclear division of alarm levels and overlapping response areas. It is difficult to accurately identify the actual operating status of each section, resulting in redundant or interference warnings. In particular, when the early warning signal is falsely triggered in a certain section, it is easy to affect the recognition results of other sections and reduce the judgment accuracy of the overall system. In addition, the existing system lacks a dynamic matching mechanism for alarm levels and operating status, and is unable to accurately adjust the alarm response according to the operating intensity and personnel density of each section. Therefore, how to reduce the risk of false triggering of early warnings in parallel maintenance of multiple sections and enhance the matching accuracy of alarm levels to the operating status of each section, so as to achieve the recognition accuracy of section-level linkage early warnings, has become a technical problem that needs to be solved urgently.

[0004] In view of this, the present invention proposes a highway maintenance and construction intelligent security management early warning method and system to solve the above problems. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a method and system for intelligent security management and early warning of highway maintenance and construction.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] In the first aspect, a method for early warning of intelligent security management for highway maintenance and construction is provided, including:

[0008] Obtain historical status data and real-time monitoring data of multi-section maintenance areas, where historical status data is used to characterize the operating status of each section during construction, and real-time monitoring data includes image connectivity data and reflection amplitude data;

[0009] Perform encoding processing on historical status data to obtain job duration features, personnel density features, and equipment frequency features, and splice the job duration features, personnel density features, and equipment frequency features to generate a status vector;

[0010] Perform connected component screening processing on image connectivity data to obtain connected region features, and perform mapping accumulation processing on reflection amplitude data to obtain reflection amplitude features;

[0011] Perform multi-scale alignment processing on the status vector and the connected region features to obtain an aligned feature sequence, and perform segmented matching processing on the aligned feature sequence and the reflection amplitude features to generate a matching matrix;

[0012] Perform difference calculation processing on the matching matrix according to a preset adjacency matrix to obtain an interference value sequence, and perform level division based on the interference value sequence to obtain a linkage warning result.

[0013] In some embodiments, the method for performing connected component screening processing on image connectivity data to obtain connected region features includes:

[0014] Perform connected component extraction processing on image connectivity data to obtain a connected component set;

[0015] Based on the centroid trajectories of each connected component in the connected component set in consecutive frame images, calculate the displacement change sequence between adjacent frames, and calculate the inter-frame fluctuation value based on the displacement change sequence as the fluctuation intensity of the corresponding connected component. Arrange the fluctuation intensities of each connected component in the order of their indices in the connected component set to form a fluctuation value sequence;

[0016] Perform threshold screening processing on the fluctuation value sequence, screen out the connected components whose fluctuation values are not greater than the preset threshold, and use the screened target connected component set as the connected region feature.

[0017] In some embodiments, the method for performing mapping accumulation processing on reflection amplitude data to obtain reflection amplitude features includes:

[0018] Based on a preset radar echo sampling time axis, divide the continuous reflection values in the reflection amplitude data into multiple frame segments by equidistant sliding windows, and keep the length of each frame the same and the inter-frame overlap ratio as the preset threshold to obtain a reflection frame sequence;

[0019] According to each frame in the reflection frame sequence, extract the corresponding two-dimensional spatial coordinate data points in each frame, and map the two-dimensional spatial coordinate data points to a grid coordinate system with a fixed resolution according to their positions to obtain a frame-level grid mapping result;

[0020] Perform spatial accumulation processing on the frame-level grid mapping results, calculate the average reflection value corresponding to each grid according to the grid index, obtain a two-dimensional reflection value matrix, and use the two-dimensional reflection value matrix as the reflection amplitude feature characterizing the spatial distribution.

[0021] In some embodiments, the method for performing multi-scale alignment processing on the state vector and the connected region feature to obtain an aligned feature sequence includes:

[0022] According to the time index and section number recorded in the operation duration feature, aggregate the operation duration feature values, personnel density feature values, and equipment frequency feature values by section number respectively, and arrange them in the order of time index within each section to construct an operation state matrix;

[0023] According to the connected component number, preset frame index, and section number, extract the stability value and area value corresponding to each connected component in each frame, construct a stability sequence and an area sequence respectively, classify the stability values and area values by section number, and sort them by frame index in each section to form a stability matrix and an area matrix;

[0024] Perform scale expansion processing on the stability matrix and the area matrix to generate a stability expansion matrix and an area expansion matrix respectively;

[0025] According to the time index recorded in the stability expansion matrix, area expansion matrix, and operation state matrix, perform index alignment processing to obtain a frame-level matching structure;

[0026] Combine the operation state vector, stability vector, and area vector in each frame-level matching structure into a single-frame feature vector according to the column vector splicing rule, and arrange all single-frame feature vectors in time order to form an aligned feature sequence.

[0027] In some embodiments, the method for performing scale expansion processing on the stability matrix and the area matrix to generate a stability expansion matrix and an area expansion matrix respectively includes:

[0028] Sort the area values of each time frame in the area matrix according to the time index, construct an area original sequence, perform window division on the area original sequence based on the preset sliding window length and sliding step, and generate an area window set;

[0029] Perform maximum value extraction processing on the area values within each window in the area window set, assign the extracted maximum value to all time frames within the window, and combine the extended results of each window in time order to generate an area expansion matrix;

[0030] Sort the stability values of each time frame in the stability matrix according to the time index, construct a stability original sequence, identify the positions of missing values or jump segments in the stability original sequence, and mark them as interpolation pending sections;

[0031] Perform linear interpolation processing on the interpolation segment to be processed to generate an interpolation stability sequence, and perform trend enhancement processing on the interpolation stability sequence to obtain a stability expansion matrix.

[0032] In some embodiments, the method of performing trend enhancement processing on the interpolation stability sequence to obtain a stability expansion matrix includes:

[0033] For each interpolation point in the interpolation stability sequence, extract the stability values of a fixed number of time frames before and after the interpolation point, combine them in chronological order into a fluctuation window sequence, and calculate the stability change direction between adjacent frames within each fluctuation window;

[0034] Perform sign consistency analysis on the stability change direction sequence in each fluctuation window, count the number of positive changes and negative changes, and calculate the direction consistency ratio as the trend consistency index for this interpolation point;

[0035] Extract the stability value of the previous frame and the stability value of the next frame corresponding to each interpolation point, calculate their absolute difference, and construct a linear combination model with the trend consistency index and this difference, and calculate the interpolation enhancement factor as the weight coefficient for interpolation adjustment;

[0036] Perform weighted fusion processing on the interpolation enhancement factor and the original interpolation value to generate an enhanced interpolation value, and arrange all the enhanced interpolation values according to the original time index to form a stability expansion matrix.

[0037] In some embodiments, the method of performing difference calculation processing on the matching matrix according to a preset adjacency matrix to obtain an interference value sequence includes:

[0038] According to the time index and segment number recorded in each cell of the matching matrix, extract the stability difference term and area difference term at the corresponding position, and form a difference frame set with these difference terms;

[0039] According to the adjacency relationship between segment numbers defined in the preset adjacency matrix, construct a segment adjacency pair set, and map each adjacency pair to the corresponding time index position in the matching matrix;

[0040] At each time index, according to the adjacent segment numbers recorded in the adjacency pair set, extract their stability difference terms and area difference terms, calculate the difference difference value between the two segments, and construct a difference term sequence;

[0041] Concatenate all the difference term sequences at each time index in a preset order to form the corresponding interference intensity frame, and arrange all the interference intensity frames in chronological order to form an interference value sequence.

[0042] In some embodiments, the method of calculating the difference difference value between two segments and constructing a difference term sequence includes:

[0043] Extract the stability difference terms and area difference terms at all time indices from the matching matrix, and organize the difference terms corresponding to all sections at each time index into a set of difference structure frames according to time;

[0044] According to each pair of adjacent section numbers recorded in the preset adjacency matrix, locate their time and position indices in the set of difference structure frames, and extract the stability difference terms and area difference terms of the adjacent pair to form a set of adjacent difference groups;

[0045] Perform difference calculation processing on the stability difference terms of the two sections in each adjacent difference group to obtain stability difference terms, and perform the same processing on the area difference terms to obtain area difference terms;

[0046] Combine each pair of stability difference terms and area difference terms into a difference vector. The difference vector contains its time index, adjacent section number, and difference value, constituting a set of difference vector frames.

[0047] In some embodiments, the method for performing difference calculation processing on the stability difference terms of the two sections in each adjacent difference group to obtain stability difference terms includes:

[0048] Obtain the stability difference terms of the two sections in the adjacent difference group, respectively mark them as the first difference and the second difference, and mark their corresponding time indices and section numbers to construct a time-aligned difference pair;

[0049] Perform amplitude normalization processing on the two stability differences in the time-aligned difference pair, use the global maximum and minimum values of the stability differences to construct a linear normalization function, and map the first difference and the second difference to the interval [0,1] to obtain a normalized stability pair;

[0050] Perform trend dissimilarity calculation processing on the normalized stability pair, calculate the direction signs of the difference sequences at consecutive time indices of the two, and count the number of times the direction signs are inconsistent to obtain a direction difference score;

[0051] Construct a weighted combination function with the absolute value of the numerical difference of the normalized stability pair and the direction difference score, and use the weighted combination function as the stability difference term.

[0052] In a second aspect, a highway maintenance construction intelligent security management warning system is provided, which is used to implement the above-mentioned highway maintenance construction intelligent security management warning method, including:

[0053] Data acquisition module: used to acquire historical state data and real-time monitoring data of a multi-section maintenance area. Among them, the historical state data is used to characterize the operation status of each section during construction, and the real-time monitoring data includes image connectivity data and reflection amplitude data;

[0054] The first processing module: It is used to perform encoding processing on historical status data to obtain job duration features, personnel density features, and equipment frequency features, and splice the job duration features, personnel density features, and equipment frequency features to generate a status vector;

[0055] The second processing module: It is used to perform connected component screening processing on image connectivity data to obtain connected region features, and perform mapping and accumulation processing on reflection amplitude data to obtain reflection amplitude features;

[0056] The third processing module: It is used to perform multi-scale alignment processing on the status vector and the connected region features to obtain an aligned feature sequence, and perform segmented matching processing on the aligned feature sequence and the reflection amplitude features to generate a matching matrix;

[0057] The early warning module: It is used to perform difference calculation processing on the matching matrix according to a preset adjacency matrix to obtain an interference value sequence, and perform level division based on the interference value sequence to obtain a linkage early warning result.

[0058] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0059] By acquiring historical status data and real-time monitoring data of multi-section maintenance areas, extracting job duration features, personnel density features, and equipment frequency features, and generating a status vector, the present invention realizes the structured representation of the operation status of each section; further performs connected component screening processing and mapping and accumulation processing on image connectivity data and reflection amplitude data respectively to obtain connected region features and reflection amplitude features, effectively identifying the visual structure and physical feedback features of the construction area; on this basis, performs multi-scale alignment processing on the status vector and the connected region features, and performs segmented matching processing on the aligned feature sequence and the reflection amplitude features to generate a matching matrix, thereby realizing the dynamic fusion matching of the early warning level and the operation status; by performing difference calculation processing on the matching matrix based on a preset adjacency matrix to obtain an interference value sequence, and performing level division accordingly, the disturbance changes between each section can be accurately identified; the finally generated linkage early warning result has a time index and a section number structure, improving the recognition accuracy of the system for the operation status of multi-section parallel maintenance operations and effectively reducing the risk of false early warning triggers. Description of the Drawings

[0060] Figure 1 It is a flow chart of a method for intelligent security management and early warning of highway maintenance construction in the present invention;

[0061] Figure 2 It is a structural diagram of an intelligent security management and early warning system for highway maintenance construction in the present invention. Detailed Embodiments

[0062] To make the objectives, technical solutions, and advantages of the present invention more clearly understood, the following further details the present invention in conjunction with specific embodiments and with reference to the accompanying drawings. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the described exemplary embodiments. However, it will be apparent to those skilled in the art that some or all of these specific details may be practiced without these specific details. In other exemplary embodiments, well-known structures are not described in detail to avoid unnecessarily obscuring the concepts of the present disclosure. It should be understood that the specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention. At the same time, the various aspects described in the embodiments can be combined arbitrarily without conflict.

[0063] Embodiment 1

[0064] Please refer to Figure 1 As shown, this embodiment publicly provides a smart security management warning method for highway maintenance construction, including:

[0065] S10: Obtain the historical status data and real-time monitoring data of multiple-section maintenance areas. Among them, the historical status data is used to represent the operation status of each section during construction, and the real-time monitoring data includes image connectivity data and reflection amplitude data;

[0066] In this embodiment, the multiple-section maintenance area refers to an operation scope composed of two or more relatively independent operation sections during road maintenance or construction. There are differences in the spatial location, operation tasks, construction progress, or equipment layout of each section. Usually, it is divided based on road segments, geographical numbers, or operation plans to support parallel construction or distributed status monitoring tasks.

[0067] For example: During the night maintenance operation of the north-south expressway in a certain city, the section from 1 km to 3 km of the southbound main lane is divided into two operation sections, A and B. Section A mainly performs road surface milling and paving, while section B performs guardrail replacement and signal light adjustment. Although the operation times of the two overlap, the operation contents and operation teams are different, belonging to a multiple-section maintenance area with spatial independence and functional heterogeneity.

[0068] It should be noted that the historical status data includes construction duration data, personnel distribution data, and equipment operation frequency data, which are used to reflect the operation intensity, resource allocation, and equipment operation rhythm of each section during construction, providing basic feature support for subsequent status coding and section alignment processing. The above data can be obtained through historical construction logs, positioning and punching records, and equipment operation records, and have structured time series tags, which are suitable for subsequent vectorization modeling and dynamic feature analysis processing.

[0069] It is understandable that the construction duration data refers to the operation duration information calculated based on the start and end time records of each section's construction tasks, usually in minutes or hours, which reflects the concentration and time intensity of the construction tasks in this section. This data can be generated by parsing the construction plan registration form or the equipment start-stop log, and is numbered sequentially according to the construction date. The personnel distribution data refers to the spatial density statistical value generated based on the real-time positioning records of personnel during construction, which is used to characterize the number of construction personnel and the activity coverage in different sections during each period. This data is usually collected by wearable positioning terminals or the construction site attendance system. After spatial grid aggregation processing, it forms a regional density matrix for status analysis. The equipment operation frequency data refers to the number of key state changes such as startup, operation, and stop of various construction equipment in each section during construction, which is used to quantify the activity level and operation load of the construction equipment in this section. This data can be extracted from the equipment operation control system or the sensor record file. After establishing a mapping relationship by combining the equipment number and the operation log, a frequency statistics table indexed by equipment category and time is generated.

[0070] It should be noted that the real-time monitoring data includes image connectivity data and reflection amplitude data, which are used to reflect the current visual structure characteristics and target physical feedback characteristics of each section, providing dynamic input support for subsequent spatial area division, operation status identification, and interference elimination processing. The above data is obtained by the image acquisition devices and millimeter-wave radar equipment deployed in front of and on both sides of the construction area, and is synchronized and aligned and data-framed through preset time windows and spatial numbering rules to meet the structured modeling requirements of the current construction environment status.

[0071] It is understandable that the image connectivity data refers to the set of closed regions or contour segments identified through image segmentation and boundary tracking algorithms based on the continuous frame images obtained by the image acquisition device, which is used to depict the spatial boundary form of continuous obstacles, warning signs, or personnel gathering areas at the construction site. After image area coding processing, this data generates a connectivity structure matrix suitable for section-level spatial feature alignment.

[0072] The reflection amplitude data refers to the reflection energy intensity values detected by the millimeter-wave radar in the construction area at different azimuth angles and distance dimensions, usually expressed in the form of a normalized amplitude array, which is used to reflect the material reflection characteristics and dynamic change trends of on-site objects. After frequency band filtering and target merging processing, this data can generate a high-sensitivity reflection feature vector at the section level for subsequent section matching and grade determination processing calls.

[0073] S20: Perform coding processing on the historical state data to obtain the operation duration feature, personnel density feature, and equipment frequency feature, and splice the operation duration feature, personnel density feature, and equipment frequency feature to generate a state vector;

[0074] In this embodiment, the historical status data includes construction duration data, personnel distribution data and equipment operation frequency data. The method of performing encoding processing on the historical status data to obtain operation duration characteristics, personnel density characteristics and equipment frequency characteristics includes:

[0075] Perform time difference calculation on the construction duration data, extract the time difference between the start and end time of construction in each section, and obtain the operation duration characteristics;

[0076] Perform spatial grid statistical processing on the personnel distribution data, count the number of positioning points in each section within a unit time, and obtain the personnel density characteristics;

[0077] The state change counting process is performed on the equipment operation frequency data, and the number of operation state switching of each type of equipment per unit time in each section is counted to obtain the equipment frequency characteristics.

[0078] The operation duration features, personnel density features and equipment frequency features are spliced ​​together to generate a state vector. The feature-level horizontal splicing method can be used to arrange the three types of feature values ​​of each section in the same time period in sequence according to a preset order to form a feature vector sequence with consistent dimensions as the state vector.

[0079] It should be noted that the time difference calculation is performed on the construction duration data, specifically: according to the start time and end time corresponding to the construction task of each section, the difference between the two is calculated to obtain the operation duration characteristics. This processing method is applicable to the timestamp information recorded by the construction schedule, progress management system or equipment start and stop log. It is a time interval calculation method known in the art and has the characteristics of clear structure and definite logic.

[0080] Perform spatial grid statistical processing on the personnel distribution data, specifically: divide the construction site into preset 2m×2m spatial grid areas, count the number of positioning points appearing in each grid within a unit time interval, and aggregate them into the corresponding construction section. The cumulative number of positioning points in each section per unit time is the personnel density feature of the section. This method needs to be based on the spatial coordinate data collected by the positioning punch-in device, combined with the mapping relationship between the section number and the grid. The spatial grid division can use a subdivision method based on the regional bounding box projection to ensure the aggregation accuracy within each section.

[0081] Perform state change counting processing on the equipment operation frequency data, specifically: analyze the operation state record sequence of each type of construction equipment in each section (for example: on → off → on → off), use "state switching" as the counting unit, and count the number of state changes that occur within a unit time. For example, if a device experiences three state change events of start → stop → restart within 15 minutes, the frequency count value of this device is 2 times (counting from off to on as 1 time, and from on to off as 1 time). The above operation state records can be extracted from the device operation control system, sensors, or vehicle CAN bus records, and it is necessary to ensure that the state records have time tags and section positioning tags to support spatio-temporal joint statistics.

[0082] S30: Perform connected component screening processing on the image connectivity data to obtain connected region features, and perform mapping accumulation processing on the reflection amplitude data to obtain reflection amplitude features;

[0083] In this embodiment, the method for performing connected component screening processing on the image connectivity data to obtain connected region features includes:

[0084] Perform connected component extraction processing on the image connectivity data to obtain a set of connected components;

[0085] Based on the centroid trajectories of each connected component in consecutive frame images in the set of connected components, calculate the displacement change sequence between adjacent frames, and calculate the inter-frame fluctuation value based on the displacement change sequence as the fluctuation intensity of the corresponding connected component. Arrange the fluctuation intensities of each connected component in the order of their indices in the set of connected components to form a fluctuation value sequence;

[0086] Perform threshold screening processing on the fluctuation value sequence, screen out the connected components with fluctuation values not greater than the preset threshold, and use the screened set of target connected components as the connected region features.

[0087] In this embodiment, performing connected component extraction processing on the image connectivity data means extracting the regions with pixel connectivity characteristics in the image as independent image segments. This processing belongs to the basic operations in image analysis. It should be noted that to achieve dynamic recognition of regional stability, further extract the centroid trajectories of each connected component in consecutive frame images, and construct a displacement change sequence by calculating the centroid displacement between adjacent frames. This trajectory information is used to quantify the drift characteristics of the connected component in the time dimension, thereby providing a basic support for judging whether it is a structural construction area.

[0088] It can be understood that to accurately reflect the stability of the connected components in the time series, the inter-frame fluctuation value is calculated based on the displacement change sequence. Usually, the fluctuation intensity of the connected component is obtained in the form of a sliding window standard deviation. The lower the fluctuation value, the more stable it is. This indicator helps to eliminate false interferences caused by non-operating targets such as personnel flow and temporary occlusion. Further, after combining the fluctuation values of all connected components into a fluctuation value sequence, a screening operation is performed based on a set stability threshold, and only the connected components with fluctuation values not greater than the threshold are retained as the final connected region features. This processing ensures that the subsequent early warning model makes response judgments only based on the stable and structurally significant regions in the image.

[0089] The methods for performing mapping accumulation processing on the reflection amplitude data to obtain the reflection amplitude features include:

[0090] Based on a preset radar echo sampling time axis, the continuous reflection values in the reflection amplitude data are divided into multiple frame segments by an equally spaced sliding window, and the length of each frame is kept consistent, and the inter-frame overlap ratio is a preset threshold to obtain a reflection frame sequence;

[0091] According to each frame in the reflection frame sequence, the corresponding two-dimensional spatial coordinate data points in each frame are extracted, and the two-dimensional spatial coordinate data points are mapped to a grid coordinate system with a fixed resolution according to their positions to obtain a frame-level grid mapping result;

[0092] Perform spatial accumulation processing on the frame-level grid mapping result, calculate the average reflection value corresponding to each grid according to the grid index, obtain a two-dimensional reflection value matrix, and use the two-dimensional reflection value matrix as the reflection amplitude feature representing the spatial distribution.

[0093] In this embodiment, to extract stable spatial reflection features, the radar echo reflection amplitude data is processed by sliding window division, specifically including: taking the preset sampling time axis as a reference, dividing the continuous reflection data into multiple frame segments by an equally spaced sliding window. During the division process, the length of each frame is kept consistent, and there is a set ratio of time overlap between frames to enhance the continuity and redundancy features of the reflection signal in time. The purpose of this processing is to ensure that the subsequent spatial mapping can construct a consistent spatial view based on time segments and avoid the error influence caused by instantaneous mutations.

[0094] It can be understood that for the reflection amplitude data within each frame segment, the corresponding two-dimensional spatial coordinate points are extracted and mapped to a preset fixed-resolution grid coordinate system according to their positions to construct a frame-level grid mapping result. This mapping process converts the unstructured original radar data into a regular spatial grid form, enabling the spatial distribution between different time frames to be accumulated and compared in a unified reference system. The grid resolution is uniformly 2 meters × 2 meters, which is consistent with the spatial grid of the personnel distribution data.

[0095] Further, perform spatial accumulation processing on all frame-level grid mapping results. Specifically, at the same grid index, count the reflection amplitudes of the corresponding grids in all frames, calculate the average reflection value of the grid, and construct a two-dimensional reflection value matrix, which is used to characterize the overall reflection intensity distribution state of the entire construction area under the current time window. The grid area with a higher reflection amplitude usually indicates the existence of hard structures, equipment, or obstacles, providing key physical inputs for subsequent spatial state recognition and regional linkage analysis.

[0096] S40: Perform multi-scale alignment processing on the state vector and the connected region features to obtain an aligned feature sequence, and perform segmented matching processing on the aligned feature sequence and the reflection amplitude features to generate a matching matrix;

[0097] The method of performing multi-scale alignment processing on the state vector and the connected region features to obtain an aligned feature sequence includes:

[0098] According to the time index and section number recorded in the operation duration feature, aggregate the operation duration feature values, personnel density feature values, and equipment frequency feature values by section number respectively, and arrange them in the order of the time index within each section to construct an operation state matrix;

[0099] According to the connected component number, preset frame index, and section number, extract the stability value and area value corresponding to each connected component in each frame, construct a stability sequence and an area sequence respectively, classify the stability values and area values by section number, and sort them in the order of the frame index in each section to form a stability matrix and an area matrix;

[0100] Perform scale expansion processing on the stability matrix and the area matrix to generate a stability expansion matrix and an area expansion matrix respectively;

[0101] According to the time index recorded in the stability expansion matrix, area expansion matrix, and operation state matrix, perform index alignment processing to obtain a frame-level matching structure;

[0102] Combine the operation state vector, stability vector, and area vector in each frame-level matching structure into a single-frame feature vector according to the column vector splicing rule, and arrange all single-frame feature vectors in chronological order to form an aligned feature sequence.

[0103] It can be understood that the construction of the operation state matrix means grouping the operation duration feature values, personnel density feature values, and equipment frequency feature values according to their corresponding section numbers, arranging them in the order of the time index within each section, splicing the three types of feature values at the same time point into a state vector, and forming an operation state matrix to reflect the construction states of each section at different time points.

[0104] It should be noted that the stability matrix and the area matrix extract the stability value and the area value of the connected component according to the connected component number, the frame index, and the section number respectively, and are organized into a two-dimensional matrix in the order of sections and time, representing the characteristic changes of each structural region in the image at different times. Scale expansion processing is performed on the stability matrix and the area matrix to enhance their temporal continuity and structural expression ability. Among them, the area matrix improves the response ability to large obstacle regions through maximum expansion of the sliding window, and the stability matrix repairs jumps and missing segments through interpolation completion and trend enhancement processing to construct a stability expansion matrix with consistent trends.

[0105] It should be noted that before the index alignment process, dynamic time warping (DTW) alignment is performed on the timestamps of the job status matrix, the stability expansion matrix, and the area expansion matrix to eliminate the clock deviation of multi-source data acquisition. The index alignment process is used to ensure the consistency of the time indexes of the job status matrix, the area expansion matrix, and the stability expansion matrix. Only the frames with time alignment in the three groups of matrices are retained to construct a frame-level matching structure. Finally, in the way of column vector splicing, the job status vector, the stability vector, and the area vector in each frame are spliced into a single-frame feature vector and arranged in chronological order to form an aligned feature sequence. This splicing operation is a conventional processing method in the art and will not be elaborated here.

[0106] The methods for performing scale expansion processing on the stability matrix and the area matrix to generate the stability expansion matrix and the area expansion matrix respectively include:

[0107] Sort the area values of each time frame in the area matrix according to the time index to construct an original area sequence, perform window division on the original area sequence based on the preset sliding window length and sliding step, and generate a set of area windows;

[0108] Perform maximum value extraction processing on the area values within each window in the set of area windows, assign the extracted maximum value to all time frames within the window, and combine the extended results of each window in chronological order to generate the area expansion matrix;

[0109] Sort the stability values of each time frame in the stability matrix according to the time index to construct an original stability sequence, identify the positions with missing values or jump segments in the original stability sequence, and mark them as interpolation segments to be processed;

[0110] Perform linear interpolation processing on the interpolation segments to be processed to generate an interpolation stability sequence, and perform trend enhancement processing on the interpolation stability sequence to obtain the stability expansion matrix.

[0111] It can be understood that the construction of the area expansion matrix is based on the time series data in the area matrix. First, the area values of each frame in the area matrix are arranged according to the time index to form the original area sequence. Then, according to the preset sliding window length and sliding step, the original sequence is window-divided to obtain a set of area windows for multiple time periods.

[0112] It should be noted that to enhance the representativeness of the area in the time dimension and the response ability to key regions, the maximum value extraction process is performed in each area window, and the maximum value is overwritten and assigned to all time frames within the window to form an enhanced area value sequence. This sequence is reorganized in chronological order, which is the area expansion matrix. This processing logic can strengthen the recognition effect of large connected component regions and reduce the unstable influence caused by small segment interference. The construction of the stability expansion matrix includes two stages. First, the stability matrix is sorted according to the time index to obtain the original stability sequence, and the regions with missing values or jump segments are detected and marked as interpolation pending processing sections. Linear interpolation is used to complete the continuity filling of these regions to generate the interpolated stability sequence.

[0113] It should be noted that during the window sliding process of the area expansion matrix and the stability expansion matrix, to avoid logical jumps or null value misalignments caused by interpolation and maximum value overwriting operations at the time series boundary, in this embodiment, at the frame segments near the boundary of the sliding window, the window alignment zero-padding method or the mirror edge-padding method is used to perform boundary expansion to ensure that each frame can participate in the window processing completely; for the regions with interpolation missing values, if they are at the start or end segment of the time series, the adjacent value copying filling or edge interpolation based on the average value of the valid frames within the window is preferentially used to ensure the continuity and stability of the interpolated sequence in the time dimension and improve the robustness of the scale expansion structure under extreme working conditions.

[0114] It should be noted that to avoid trend distortion caused by linear interpolation, a trend enhancement processing mechanism is introduced to the interpolated stability sequence. By analyzing the stability change directions of several frames before and after the interpolation point, the consistency ratio of the fluctuation directions is statistically calculated, a trend consistency factor is constructed, and the enhancement coefficient is calculated in combination with the numerical difference between adjacent frames. Finally, the enhancement coefficient is applied to the interpolation result to form a stability expansion matrix with trend preservation characteristics. It should be added that for the missing values at the start or end segment of the time series, the first frame copying method or the last frame mirror method is used for filling to ensure interpolation continuity.

[0115] The method for performing trend enhancement processing on the interpolated stability sequence to obtain the stability expansion matrix includes:

[0116] For each interpolation point in the interpolated stability sequence, extract the stability values of a fixed number of time frames before and after the interpolation point, combine them in chronological order to form a fluctuation window sequence, and calculate the stability change direction between adjacent frames within each fluctuation window;

[0117] Perform sign consistency analysis on the stability change direction sequence in each fluctuation window, count the number of positive changes and the number of negative changes, and calculate the direction consistency ratio as the trend consistency index of this interpolation point;

[0118] Extract the stability values of the previous frame and the next frame corresponding to each interpolation point, calculate their absolute difference, and construct a linear combination model with the trend consistency index and this difference to calculate the interpolation enhancement factor as the weight coefficient for interpolation adjustment;

[0119] Perform weighted fusion processing on the interpolation enhancement factor and the original interpolation value to generate an enhanced interpolation value, and arrange all the enhanced interpolation values according to the original time index to form a stability extension matrix.

[0120] It can be understood that to achieve the structural preservation of the stability change trend in the interpolation result, for each interpolation point in the interpolation stability sequence, extract a fixed number of time frames before and after it to construct a set of stability values including this point in the time neighborhood, and form a sequence of fluctuation windows in chronological order. Calculate the difference direction of the stability values between adjacent time frames in each window to represent the positive and negative trends of local changes. To identify the consistency of the fluctuation direction in the region where the interpolation point is located, perform sign consistency analysis on the above change direction sequence, count the number of positive changes and the number of negative changes respectively, and calculate their proportional relationship to obtain the trend consistency index of this interpolation point. This index is used to evaluate the stability degree of the local fluctuation trend and is the logical basis for subsequent interpolation adjustment.

[0121] Furthermore, to improve the adaptability of the interpolation to the adjacent change amplitude, extract the stability values of the previous frame and the next frame of each interpolation point, calculate their absolute difference, and construct a linear combination function with this difference and the trend consistency index to calculate the interpolation enhancement factor as the weight coefficient for adjusting the interpolation size, which is used to enhance the coupling relationship between the interpolation point and the local trend. Finally, perform weighted fusion processing on the interpolation enhancement factor and the initial interpolation to generate an enhanced interpolation. The stability values of all interpolation points after enhancement are arranged in chronological order and combined into a stability extension matrix. This processing method ensures the fluctuation coherence and local change consistency of the interpolation structure in the time series, and improves the overall trend accuracy of the stability expression.

[0122] The method of generating a matching matrix by performing segmented matching processing on the aligned feature sequence and the reflection amplitude feature includes:

[0123] According to the time index in each single-frame feature vector, extract the reflection value at the corresponding moment from the reflection amplitude data, and form a reflection value sequence with all the reflection values in chronological order;

[0124] For each single-frame feature vector, extract the corresponding time index and section number, and locate the reflection value at the same time and the same section in the reflection value sequence to construct a matching data pair corresponding to the feature vector;

[0125] For each matching data pair, extract the stability value and area value in the feature vector, and perform difference calculations with the matched reflection value respectively to obtain the stability difference and area difference;

[0126] Organize all the stability differences and area differences into a two-dimensional matrix according to the time index and section number, and use the two-dimensional matrix as the matching matrix.

[0127] It can be understood that to ensure the time consistency between the aligned feature sequence and the reflection amplitude feature, it is necessary to extract the reflection value at the corresponding moment from the reflection amplitude data based on the time index recorded in the single-frame feature vector, and arrange the extraction results in the order of the time index to form a reflection value sequence. This processing ensures that all reflection values and state data have a unified time axis basis, which is a prerequisite for the subsequent construction of the matching structure. Further, it is necessary to extract the corresponding time index and section number for each single-frame feature vector, and search for the reflection value at the same time and section in the reflection value sequence to construct a matching data pair corresponding to the feature vector. Through the combined positioning operation of the time index and section number, the spatial and time accuracy of the reflection value matching can be ensured, avoiding mis-matching between different sections or non-synchronized frames. The construction result of the matching data pair will be used as the input for the difference calculation.

[0128] It should be noted that it is necessary to extract the stability value and area value of this frame from each group of matching data pairs, and perform numerical difference calculations on the two with the corresponding reflection value respectively to obtain the stability difference and area difference. This processing can be used to measure the deviation degree between the visual feature and the radar physical feature. The smaller the numerical difference, the stronger the perceptual structure consistency of the two types of features at this moment, which is the basic quantitative index for subsequent linkage determination or anomaly detection. Organize all the obtained stability differences and area differences into a two-dimensional matrix in a dual-indexing manner according to the time index and section number to form a matching matrix. Each row in this matrix corresponds to a specific spatio-temporal point, and each column is the stability difference and area difference respectively. This structure not only retains the original spatial segmentation characteristics but also integrates the offset information between the visual and radar features, which can be used as the input basis for subsequent interference degree estimation or level determination modules.

[0129] S50: Perform difference calculation processing on the matching matrix according to the preset adjacency matrix to obtain an interference value sequence, and perform level division based on the interference value sequence to obtain a linkage warning result.

[0130] The method of performing difference calculation processing on the matching matrix according to the preset adjacency matrix to obtain an interference value sequence includes:

[0131] According to the time index and section number recorded in each cell of the matching matrix, extract the stability difference term and area difference term at the corresponding position, and form a set of difference frames with these difference terms;

[0132] According to the adjacency relationship between section numbers defined in the preset adjacency matrix, construct a set of section adjacency pairs, and map each adjacency pair to the corresponding time index position in the matching matrix;

[0133] At each time index, according to the adjacent section numbers recorded in the set of adjacency pairs, extract their stability difference terms and area difference terms, calculate the difference value between the two sections, and construct a sequence of difference terms;

[0134] Concatenate all the sequences of difference terms at each time index in the preset order to form the corresponding interference intensity frame, and arrange all the interference intensity frames in time order to form an interference value sequence.

[0135] In this embodiment, the preset adjacency matrix is calculated based on the coordinates of the construction section geographic information system (GIS). If the boundary distance between two sections is ≤ 5 meters, it is marked as an adjacency relationship (value is 1), otherwise it is 0. The matching matrix is a two-dimensional structure constructed according to the difference between the alignment feature sequence and the reflection amplitude feature, organized by time index and section number. Each cell records the stability difference term and area difference term respectively, serving as the basic input for constructing the disturbance relationship. The adjacency matrix represents whether there is a physical adjacency or operation linkage relationship between construction sections, and its structure is usually a Boolean matrix, where the value of 1 indicates that two sections are adjacent. This matrix provides a structural basis for extracting the disturbance path in the spatial dimension later.

[0136] In this embodiment, the combined processing of the matching matrix and the adjacency matrix includes the following steps: First, extract the stability difference terms and area difference terms of all sections at each time index to form a set of difference frames; then, according to the adjacency relationship defined in the adjacency matrix, traverse each pair of adjacent sections, take out their stability difference terms at the same time point and calculate the difference between them to obtain the stability difference value; perform the same difference processing on the area difference terms to generate the area difference value. Combine the set of difference values calculated for all adjacency pairs at the same time point in order to form an interference intensity frame, and then arrange all the interference intensity frames in time dimension in sequence to generate an interference value sequence. This sequence reflects the disturbance conduction intensity between sections and is the direct basis for subsequent disturbance level determination and alarm generation.

[0137] The method for calculating the difference value between two sections and constructing a sequence of difference terms includes:

[0138] Extract the stability difference terms and area difference terms at all time indexes from the matching matrix, and organize the difference terms corresponding to all sections at each time index into a set of difference structure frames according to time;

[0139] According to each pair of adjacent section numbers recorded in the preset adjacency matrix, locate their time and position indices in the set of difference structure frames, extract the stability difference terms and area difference terms of the adjacent pairs, and form a set of adjacent difference groups;

[0140] Perform difference calculation processing on the stability difference terms of the two sections in each adjacent difference group to obtain stability difference items, and perform the same processing on the area difference terms to obtain area difference items;

[0141] Combine each pair of stability difference items and area difference items into a difference vector. The difference vector contains its time index, adjacent section number, and difference value, constituting a set of difference vector frames.

[0142] It can be understood that the construction of the difference vector in this embodiment is a key step in extracting the relationship between the stability and area differences between sections under the dual structural constraints of the matching matrix and the adjacency matrix. The purpose is to characterize the degree of perturbation change between spatially adjacent sections under the same time index. Extract all recorded time indices from the matching matrix and traverse them in chronological order. Under each time index, extract the stability difference terms and area difference terms corresponding to all sections, sort them by section number, and form a set of difference structure frames. The set of difference structure frames is used to identify the perturbation value structure of all spatial sections at each moment, providing a spatio-temporally complete input source for subsequent difference calculations. According to the defined adjacency relationship in the preset adjacency matrix, enumerate all combinations of section adjacent pairs, and locate the stability difference terms and area difference terms corresponding to the two sections involved in the adjacent pair in the set of difference structure frames according to the time index, and construct a set of adjacent difference groups. The set of adjacent difference groups retains the time, number, and two perturbation values of each group of adjacent sections in terms of structure.

[0143] It should be noted that in each group of adjacent difference groups, perform a numerical subtraction operation on the stability difference terms corresponding to the two sections respectively to obtain stability difference items. Similarly, perform a difference calculation on the area difference terms to obtain area difference items. The above operations can be formalized as: , where is the stability difference item, is the area difference item, are the first stability difference term and the second stability difference term respectively, are the first area difference term and the second area difference term respectively. Finally, combine the stability difference item and the area difference item to generate a difference vector containing the time index, adjacent section number, and two difference values, and stack all difference vectors in chronological order of the time index to form a set of difference vector frames. The set of difference vector frames serves as a structured input for subsequent interference intensity extraction and dynamic warning processing.

[0144] Performing difference calculation processing on the stability difference terms of two segments in each adjacent difference group to obtain the stability difference term, the method includes:

[0145] Obtain the stability difference terms of two segments in the adjacent difference group, respectively marked as the first difference and the second difference, and mark their corresponding time indices and segment numbers to construct a time-aligned difference pair;

[0146] Perform amplitude normalization processing on the two stability differences in the time-aligned difference pair, use the global maximum and minimum values of the stability differences to construct a linear normalization function, and map the first difference and the second difference to the interval [0,1] to obtain a normalized stability pair;

[0147] Perform trend dissimilarity calculation processing on the normalized stability pair, calculate the direction signs of the difference sequences of the two at consecutive time indices, and count the number of times the direction signs are inconsistent to obtain a direction difference score;

[0148] Construct a weighted combination function with the absolute value of the numerical difference of the normalized stability pair and the direction difference score, and use the weighted combination function as the stability difference term.

[0149] It can be understood that to ensure the standardization and trend response analysis of the differences between stability difference terms, first, the stability difference terms corresponding to two segments need to be extracted from the adjacent difference group, marked as the first difference and the second difference respectively, and the time index and segment number information corresponding to both are recorded, and combined into a time-aligned difference pair to ensure that subsequent processing operations perform logically consistent difference analysis on the same time dimension and adjacent structure.

[0150] It should be noted that to eliminate the interference of the fluctuation offset caused by the absolute values of the stability difference terms between different times or segments, amplitude normalization processing is performed on the time-aligned difference pair, that is, a linear normalization function is constructed based on the global maximum and minimum values of all stability difference terms as:

[0151] f(x)=(x - min) / (max - min);

[0152] Among them, f(x) represents transforming the original stability difference into a standardized value within the interval [0,1], x represents an original value in the stability difference sequence, min represents the global minimum stability difference value in the entire stability difference sequence, and max represents the global maximum stability difference value in the entire stability difference sequence.

[0153] And perform this function transformation on the first difference and the second difference respectively to generate a normalized stability pair. This processing can ensure the comparability of different segments on the numerical scale and improve the sensitivity of the difference score.

[0154] Furthermore, perform trend dissimilarity calculation on the normalized stability pair. Trend dissimilarity refers to whether the change directions of two stability difference terms are consistent in the adjacent time index sequence. The specific method is as follows: construct the stability difference change sequence of each section, extract the change direction (the symbol is +1 or -1) between each frame and the previous frame, and then perform symbol difference statistics on the change directions of the two sections at the same time point, accumulate the number of frames with inconsistent symbols to form a direction difference score. This score reflects whether the disturbance trends between the two sections are consistent and has the capabilities of direction sensitivity and stability characterization. Finally, to uniformly measure the amplitude difference of stability values and the direction difference of disturbances, calculate the absolute value of the numerical difference for the normalized stability pair, and perform weighted combination processing on the obtained result and the direction difference score. The form of the combination function can be:

[0155] Diff_score = α × abs( Dir_difff;

[0156] where Diff_score is the difference score, abs( ) is the absolute value of the numerical difference between the normalized stability difference terms of two adjacent sections at the same time index, Dir_difff is the direction difference score, and are the first stability difference term and the second stability difference term respectively, α is the amplitude difference weight coefficient, and β is the direction difference weight coefficient.

[0157] In this embodiment, based on the interference value sequence, perform level division to obtain the linkage warning result, which means quantifying and comparing the disturbance intensities between different sections at different times and performing threshold division processing on the basis of the constructed interference value sequence, so as to generate a warning label set that can be used for hierarchical response. In the interference value sequence, each interference intensity frame contains the stability difference terms and area difference terms of all adjacent section combinations corresponding to the time index. This structure has clear time dimension and spatial structure characteristics, and can support fine-grained perception and warning recognition of disturbance trends. To avoid misjudgment caused by instantaneous anomalies, this embodiment preferentially performs time-series sliding window smoothing processing on each difference term in the interference value sequence, and constructs an interference moving average sequence using known smoothing algorithms such as weighted average or exponential weighting to ensure the stability and time consistency of the change trend of the disturbance intensity, providing a reliable input basis for level determination.

[0158] It can be understood that the core of level division is to discretize continuous disturbance values into level intervals. Therefore, this embodiment further introduces a multi-threshold interval determination mechanism, and comprehensively scores the corresponding stability difference values and area difference values in each difference vector frame of the interference value sequence. The scoring process uses the following weighted scoring function: where ΔS represents the stability difference term, and ΔA represents the area difference term, and are preset weight factors, which respectively control the influence degrees of two types of difference items in the final score. According to the score result, the score value is compared with the pre-calibrated multi-level threshold, and is respectively divided into four levels of "no disturbance", "mild disturbance", "moderate disturbance" and "severe disturbance". Each level corresponds to different early warning response strategies, such as enhancing the alarm priority, starting linkage control mechanisms such as neighboring cell linkage barriers, etc., to ensure that the system can execute reasonable control feedback and on-site alarm linkage operations according to the real-time interference degree.

[0159] It should be noted that, in order to avoid ambiguity in numerical stability and calculation implementation of the scoring function or trend enhancement function, in this embodiment, the weighted parameters in the scoring function can be automatically learned by the least mean square or cross-entropy optimization method based on the historical disturbance data set in the training stage, or can be obtained by manually setting and repeatedly experimenting to obtain empirical values; for the construction process of the trend consistency index and interpolation enhancement factor involved in the trend enhancement function, existing fluctuation detection algorithms and trend prediction models (such as moving average trend recognition, short-term trend sliding window difference analysis) can be used to avoid affecting the clarity of the implementation path due to the lack of explicit specific function expressions.

[0160] Embodiment 2

[0161] Please refer to Figure 2 as shown. Based on the same inventive concept, this embodiment discloses and provides. For the details not described in this embodiment, please refer to the relevant parts in Embodiment 1. The system includes:

[0162] Data acquisition module: used to acquire the historical state data and real-time monitoring data of the multi-section maintenance area. Among them, the historical state data is used to represent the operation state of each section during construction, and the real-time monitoring data includes image connectivity data and reflection amplitude data;

[0163] The first processing module: used to perform encoding processing on the historical state data to obtain the operation duration feature, personnel density feature and equipment frequency feature, and splice the operation duration feature, personnel density feature and equipment frequency feature to generate a state vector;

[0164] The second processing module: used to perform connected component screening processing on the image connectivity data to obtain the connected region feature, and perform mapping accumulation processing on the reflection amplitude data to obtain the reflection amplitude feature;

[0165] The third processing module: used to perform multi-scale alignment processing on the state vector and the connected region feature to obtain an alignment feature sequence, and perform segmented matching processing on the alignment feature sequence and the reflection amplitude feature to generate a matching matrix;

[0166] Early warning module: configured to perform difference calculation processing on a matching matrix according to a preset adjacency matrix to obtain a sequence of interference values, and perform level division based on the sequence of interference values to obtain a linkage early warning result.

[0167] The detailed description set forth above in connection with the accompanying drawings describes exemplary embodiments and not all embodiments that may be implemented or that fall within the scope of the claims. The terms "exemplary" and "exemplification" as used herein mean "serving as an example, instance, or illustration" and do not mean "superior to or better than other examples".

[0168] References throughout this specification to "one embodiment" or "an embodiment" mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, the appearances of these phrases are not necessarily all referring to the same embodiment. Further, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0169] It should also be noted that these embodiments may be described as processes depicted as flowcharts, structure diagrams, or block diagrams. Although a flowchart may describe the operations as a sequential process, many of the operations can be performed in parallel or concurrently. In addition, the order of these operations may be rearranged.

Claims

1. A smart security management and early warning method for highway maintenance construction, characterized in that, Including: Obtain the historical status data and real-time monitoring data of the multi-segment maintenance area. Among them, the historical status data is used to characterize the operation status of each segment during construction, and the real-time monitoring data includes image connectivity data and reflection amplitude data; Perform encoding processing on the historical status data to obtain the operation duration feature, personnel density feature, and equipment frequency feature, and splice the operation duration feature, personnel density feature, and equipment frequency feature to generate a status vector; Perform connected component screening processing on the image connectivity data to obtain the connected region feature, and perform mapping accumulation processing on the reflection amplitude data to obtain the reflection amplitude feature; Perform multi-scale alignment processing on the status vector and the connected region feature to obtain an aligned feature sequence, and perform segmented matching processing on the aligned feature sequence and the reflection amplitude feature to generate a matching matrix; Perform difference calculation processing on the matching matrix according to the preset adjacency matrix to obtain an interference value sequence, and perform level division based on the interference value sequence to obtain the linkage warning result.

2. The intelligent security management warning method for highway maintenance construction according to claim 1, characterized in that The method for performing connected component screening processing on the image connectivity data to obtain the connected region feature includes: Perform connected component extraction processing on the image connectivity data to obtain a connected component set; Based on the centroid trajectories of each connected component in the connected component set in consecutive frame images, calculate the displacement change sequence between adjacent frames, and calculate the inter-frame fluctuation value based on the displacement change sequence as the fluctuation intensity of the corresponding connected component. Arrange the fluctuation intensities of each connected component in the index order in the connected component set to form a fluctuation value sequence; Perform threshold screening processing on the fluctuation value sequence, screen out the connected components with fluctuation values not greater than the preset threshold, and use the screened target connected component set as the connected region feature.

3. A smart security management warning method for highway maintenance construction according to claim 1, characterized in that, The method for performing mapping accumulation processing on the reflection amplitude data to obtain the reflection amplitude feature includes: Based on the preset radar echo sampling time axis, divide the continuous reflection values in the reflection amplitude data into multiple frame segments by equidistant sliding windows, and keep the length of each frame the same and the inter-frame overlap ratio as the preset threshold to obtain a reflection frame sequence; According to each frame in the reflection frame sequence, extract the corresponding two-dimensional spatial coordinate data points in each frame, and map the two-dimensional spatial coordinate data points to a grid coordinate system with a fixed resolution according to their positions to obtain a frame-level grid mapping result; Perform spatial accumulation processing on the frame-level grid mapping result, calculate the average reflection value corresponding to each grid according to the grid index, obtain a two-dimensional reflection value matrix, and use the two-dimensional reflection value matrix as the reflection amplitude feature representing the spatial distribution.

4. A smart security management and early warning method for highway maintenance construction according to claim 2, characterized in that, The method for performing multi-scale alignment processing on the status vector and the connected region feature to obtain an aligned feature sequence includes: According to the time index and segment number recorded in the operation duration feature, aggregate the operation duration feature values, personnel density feature values, and equipment frequency feature values by segment number respectively, and arrange them in the order of time index within each segment to construct an operation status matrix; According to the connected component number, preset frame index, and segment number, extract the stability value and area value corresponding to each connected component in each frame, respectively construct a stability sequence and an area sequence, classify the stability value and area value by segment number, and sort them in the frame index order in each segment to form a stability matrix and an area matrix; Perform a scaling process on the stability matrix and the area matrix to generate a stability extended matrix and an area extended matrix respectively; According to the time indices recorded in the stability extended matrix, the area extended matrix, and the job status matrix, perform an index alignment process to obtain a frame-level matching structure; Combine the job status vector, the stability vector, and the area vector in each frame-level matching structure into a single-frame feature vector according to the column vector concatenation rule, and arrange all the single-frame feature vectors in chronological order to form an aligned feature sequence.

5. A smart security management warning method for highway maintenance construction according to claim 4, characterized in that The method of performing a scaling process on the stability matrix and the area matrix to generate a stability extended matrix and an area extended matrix respectively includes: Sort the area values of each time frame in the area matrix according to the time index, construct an area original sequence, perform window partitioning on the area original sequence based on a preset sliding window length and a sliding step size, and generate an area window set; Perform a maximum value extraction process on the area values within each window in the area window set, assign the extracted maximum value to all time frames within the window, and combine the extended results of each window in chronological order to generate an area extended matrix; Sort the stability values of each time frame in the stability matrix according to the time index, construct a stability original sequence, identify the positions with missing values or jump segments in the stability original sequence, and mark them as interpolation segments to be processed; Perform a linear interpolation process on the interpolation segments to be processed to generate an interpolated stability sequence, and perform a trend enhancement process on the interpolated stability sequence to obtain a stability extended matrix.

6. The intelligent security management and early warning method for highway maintenance construction according to claim 5, characterized in that, The method of performing a trend enhancement process on the interpolated stability sequence to obtain a stability extended matrix includes: For each interpolation point in the interpolated stability sequence, extract the stability values of a fixed number of time frames before and after the interpolation point, combine them into a fluctuation window sequence in chronological order, and calculate the stability change direction between adjacent frames within each fluctuation window; Perform a sign consistency analysis on the stability change direction sequence within each fluctuation window, count the number of positive changes and the number of negative changes, and calculate the direction consistency ratio as the trend consistency index for this interpolation point; Extract the stability value of the previous frame and the stability value of the next frame corresponding to each interpolation point, calculate their absolute difference, and construct a linear combination model with the trend consistency index and this difference to calculate an interpolation enhancement factor as the weight coefficient for interpolation adjustment; Perform a weighted fusion process on the interpolation enhancement factor and the original interpolation value to generate an enhanced interpolation value, arrange all the enhanced interpolation values according to the original time index, and combine them into a stability extended matrix.

7. A method for intelligent security management and early warning in highway maintenance construction according to claim 6, characterized in that, The method of performing a difference calculation process on the matching matrix according to a preset adjacency matrix to obtain a sequence of interference values includes: According to the time index and section number recorded in each cell of the matching matrix, extract the stability difference term and the area difference term at the corresponding position, and form a difference frame set with these difference terms; According to the adjacency relationship between section numbers defined in the preset adjacency matrix, construct a set of section adjacency pairs, and map each adjacency pair to the corresponding time index position in the matching matrix; At each time index, extract the stability difference term and the area difference term according to the adjacent section numbers recorded in the set of section adjacency pairs, calculate the difference difference value between the two sections, and construct a sequence of difference terms; Concatenate all the difference item sequences at each time index in the preset order to form the corresponding interference intensity frame, and arrange all the interference intensity frames in chronological order to form an interference value sequence.

8. A method for intelligent security management and early warning of highway maintenance construction according to claim 7, characterized in that, Calculate the difference value between two sections. The methods for constructing the difference item sequence include: Extract the stability difference items and area difference items at all time indexes from the matching matrix, and organize the difference items corresponding to all sections at each time index into a difference structure frame set according to time. According to the numbers of each pair of adjacent sections recorded in the preset adjacency matrix, locate their time and position indexes in the difference structure frame set, and extract the stability difference items and area difference items of the adjacent pair to form an adjacent difference group set. Perform difference calculation processing on the stability difference items of the two sections in each adjacent difference group to obtain stability difference items, and perform the same processing on the area difference items to obtain area difference items. Combine each pair of stability difference items and area difference items into a difference vector. The difference vector contains its time index, adjacent section number and difference value, and constitutes a difference vector frame set.

9. The intelligent security management and early warning method for highway maintenance construction according to claim 8, characterized in that, The methods for performing difference calculation processing on the stability difference items of the two sections in each adjacent difference group to obtain stability difference items include: Obtain the stability difference items of the two sections in the adjacent difference group, mark them as the first difference and the second difference respectively, and mark their corresponding time indexes and section numbers to construct a time-aligned difference pair. Perform amplitude normalization processing on the two stability differences in the time-aligned difference pair, use the global maximum and minimum values of the stability differences to construct a linear normalization function, and map the first difference and the second difference to the interval [0,1] to obtain a normalized stability pair. Perform trend dissimilarity calculation processing on the normalized stability pair, calculate the direction signs of the difference sequences at consecutive time indexes of the two, and count the number of inconsistent direction signs to obtain a direction difference score. Construct a weighted combination function with the absolute value of the numerical difference of the normalized stability pair and the direction difference score, and use the weighted combination function as the stability difference item.

10. A smart security management warning system for highway maintenance construction, which is used to implement a smart security management warning method for highway maintenance construction described in any one of claims 1-9, and is characterized in that, Include: Data acquisition module: used to acquire the historical state data and real-time monitoring data of the multi-section maintenance area. Among them, the historical state data is used to represent the operation status of each section during construction, and the real-time monitoring data includes image connectivity data and reflection amplitude data. First processing module: used to perform coding processing on the historical state data to obtain the operation duration feature, personnel density feature and equipment frequency feature, and splice the operation duration feature, personnel density feature and equipment frequency feature to generate a state vector. Second processing module: used to perform connected component screening processing on the image connectivity data to obtain a connected region feature, and perform mapping accumulation processing on the reflection amplitude data to obtain a reflection amplitude feature. Third processing module: used to perform multi-scale alignment processing on the state vector and the connected region feature to obtain an alignment feature sequence, and perform segmented matching processing on the alignment feature sequence and the reflection amplitude feature to generate a matching matrix. Early warning module: used to perform difference calculation processing on the matching matrix according to the preset adjacency matrix to obtain an interference value sequence, and perform level division based on the interference value sequence to obtain a linkage early warning result.

Images