Method and device for monitoring local vehicle

Abstract

The embodiment of the invention provides a method and device for monitoring a local vehicle, and the method comprises the steps: determining a monitored vehicle through querying initial monitoring data meeting a preset configuration condition; performing incremental secondary query according to a preset query period and a preset configuration condition to obtain target monitoring data; marking the target monitoring data according to a time sequence and grouping the target monitoring data according to the identified license plate numbers to obtain a grouped marking result, and obtaining an action state sequence corresponding to the action state of the monitored vehicle according to the marking result; and establishing a monitoring time window for the monitored vehicle, determining behavior characteristics of the monitored vehicle based on the marking result and the action state in the monitoring time window, obtaining a corresponding alarm vehicle based on the behavior characteristics, and sending the alarm vehicle to a preset terminal device. According to the method, the defects that the deployment cost is high and the localization calculation efficiency is limited are effectively overcome, the deployment cost is remarkably reduced, and the localization calculation efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and device for monitoring local vehicles. Background Art

[0002] Monitoring large-scale mobile targets usually relies on a distributed computing framework to implement analysis algorithms. This requires deploying an independent big data platform, which results in high server resource usage, high O&M complexity, and the need for additional storage space to support data calculations. For scenarios with non-complex algorithms, the resource overhead of the distributed architecture does not match actual needs, resulting in cost waste.

[0003] In existing technologies, efficiency can be improved by optimizing distributed cluster resource allocation or using streaming computing engines. However, it is still impossible to avoid dependence on the underlying platform, and it is difficult to balance query performance and algorithm flexibility under large data volumes. Summary of the Invention

[0004] In response to the problems in the existing technology, the present application provides a method and device for monitoring local vehicles, which can effectively solve the shortcomings of traditional technologies such as high deployment costs and limitations in localized computing efficiency, significantly reduce deployment costs, and improve localized computing efficiency.

[0005] In order to solve at least one of the above problems, the present application provides the following technical solutions: In a first aspect, the present application provides a method for monitoring a local vehicle, comprising: receiving a query instruction, querying initial monitoring data that meets preset configuration conditions based on the query instruction, inputting the initial monitoring data into a trained spatiotemporal clustering model, and extracting target spatial clusters that meet a preset density threshold using the trained spatiotemporal clustering model, wherein the preset configuration conditions include a parking range and / or preset information data within the parking range; Determine the centroid coordinates and diffusion speed of each target spatial cluster, and when the diffusion speed is greater than a preset movement threshold, generate a virtual monitoring vehicle including a centroid trajectory corresponding to the centroid coordinates, and determine the object corresponding to the virtual monitoring vehicle as the monitoring vehicle; According to the preset query cycle and preset configuration conditions, an incremental secondary query is performed on the real-time monitoring data corresponding to the monitored vehicle to obtain the target monitoring data that meets the preset configuration conditions; The target monitoring data is marked in chronological order and grouped according to the recognized license plate numbers to obtain grouped marking results. The marking results are then compared with the target monitoring data in time and space to obtain the action state sequence corresponding to the action state of the monitored vehicle. The action state includes in-bounds, out-bounds, and stable. The marking results and the action status corresponding to the marking results are stored in a preset cache database, a monitoring time window is established for the monitored vehicle, the behavioral characteristics of the monitored vehicle are determined based on the marking results and the action status within the monitoring time window, and the monitored vehicles that meet the preset alarm conditions are screened based on the behavioral characteristics, and the corresponding alarm vehicles are obtained. The alarm vehicles are then sent to the preset terminal device so that the alarm vehicles can be processed through the terminal device.

[0006] Furthermore, the method further includes: receiving an initial query cycle and a preset buffer time, and obtaining a previous query time; If there is a previous query time, the sum of the initial query period and the preset buffer time is determined as the preset query period, and the previous query time is used as the starting query time to perform an incremental secondary query on the real-time monitoring data corresponding to the time period of the preset query period; When there is no previous query time, the initial query period is determined as the preset query period, the starting query time is determined based on the query instruction, and an incremental secondary query is performed on the real-time monitoring data corresponding to the time period of the preset query period; When it is detected that the transmission delay of the real-time monitoring data exceeds the preset buffer time, a compensation query is performed on the real-time monitoring data with the transmission delay, and the preset buffer time is adjusted.

[0007] Furthermore, the method further includes: assigning a target name to each monitored vehicle, and constructing a spatiotemporal state matrix with the target name as a primary key, wherein the row dimension of the spatiotemporal state matrix represents a time series, and the column dimension of the spatiotemporal state matrix includes the position coordinates of the monitored vehicles and the regional topological relationship; Determine the similarity between consecutive time slices in the current spatiotemporal state matrix, quantify the intensity of changes in the vehicle's action state in consecutive time slices based on the similarity metric, and insert an abnormal marker in the action state sequence when the intensity of the change exceeds a preset change threshold, where each time slice represents a row in the spatiotemporal state matrix.

[0008] Furthermore, the method further includes: determining the timestamp field in the marking result as a primary index key, and storing the marking result in a columnar storage structure in a preset cache database based on the primary index key; The marking results corresponding to hot data that exceeds the preset validity period are stored in distributed files, and the marking results corresponding to cold data that exceeds the preset validity period are stored in the memory database.

[0009] Furthermore, the method further includes: for each monitored vehicle, based on the behavioral characteristics corresponding to the current preset query cycle and the initial monitoring data, determining that the duration of the currently monitored vehicle's stay in the current parking range exceeds the preset retention duration and the action state of the monitored vehicle is stable, determining the monitored vehicle as an alarm vehicle, and updating the corresponding stay start time of the monitored vehicle to the start query time of the next preset query cycle, wherein the stay start time is the time point at which the monitored vehicle first appears; For each monitored vehicle, based on the behavioral characteristics and initial monitoring data corresponding to the current preset query cycle, it is determined that the current monitored vehicle's stay time in the current parking range exceeds the preset stay time and the monitored vehicle's action state is out of bounds, and the corresponding stay time of the monitored vehicle is reset.

[0010] Furthermore, the method further includes: for each monitored vehicle, determining, based on the behavioral characteristics corresponding to the current preset query period and the initial monitoring data, the number of out-of-bounds times and the number of in-bounds times of the corresponding action state of the currently monitored vehicle within the current parking range; The number of out-of-bounds and in-bounds is determined as one network touch count. If the number of network touch counts within a preset network touch cycle is not less than the preset network touch count, the monitored vehicle is determined as an alarm vehicle. When it is detected that the current preset network contact cycle ends, the number of network contacts in the current preset network contact cycle is cleared, and the next preset network contact cycle is started.

[0011] In a second aspect, the present application provides a device for monitoring local vehicles, comprising: a first processing module configured to receive a query instruction, query initial monitoring data that meets preset configuration conditions based on the query instruction, input the initial monitoring data into a trained spatiotemporal clustering model, and extract target spatial clusters that meet a preset density threshold using the trained spatiotemporal clustering model, wherein the preset configuration conditions include a parking range and / or preset information data within the parking range; The second processing module is configured to determine the centroid coordinates and diffusion speed of each target spatial cluster, and when the diffusion speed is greater than a preset movement threshold, generate a virtual monitoring vehicle including a centroid trajectory corresponding to the centroid coordinates, and determine the object corresponding to the virtual monitoring vehicle as the monitoring vehicle; The third processing module is used to perform an incremental secondary query on the real-time monitoring data corresponding to the monitored vehicle according to the preset query cycle and preset configuration conditions to obtain target monitoring data that meets the preset configuration conditions; The fourth processing module is used to mark the target monitoring data in chronological order and group them according to the recognized license plate numbers to obtain grouped marking results, compare the marking results with the target monitoring data in time and space, and obtain an action state sequence corresponding to the action state of the monitored vehicle, where the action state includes in-bounds, out-of-bounds, and stable; The fifth processing module is used to store the marking results and the action status corresponding to the marking results through a preset cache database, establish a monitoring time window for the monitored vehicle, determine the behavioral characteristics of the monitored vehicle based on the marking results and action status within the monitoring time window, and screen the monitored vehicles that meet the preset alarm conditions based on the behavioral characteristics, obtain the corresponding alarm vehicle, and send the alarm vehicle to a preset terminal device so that the alarm vehicle can be processed through the terminal device.

[0012] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method for monitoring local vehicles when executing the program.

[0013] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method for monitoring local vehicles when executed by a processor.

[0014] In a fifth aspect, the present application provides a computer program product, comprising a computer program / instruction, which implements the steps of the method for monitoring local vehicles when executed by a processor.

[0015] As can be seen from the above technical solution, the present application provides a method and device for monitoring local vehicles, which obtains initial monitoring data that meets preset configuration conditions by receiving query instructions, inputs the initial monitoring data into a spatiotemporal clustering model to extract target spatial clusters that meet a preset density threshold, wherein the preset configuration conditions include a parking range and / or preset information data within the parking range, determines the centroid coordinates and diffusion speed of each target spatial cluster, and when the diffusion speed is greater than a preset movement threshold, generates a virtual monitoring vehicle including a centroid trajectory corresponding to the centroid coordinates, determines the object corresponding to the virtual monitoring vehicle as a monitoring vehicle, and obtains the target monitoring vehicle through a periodic incremental query corresponding to a preset period. Data, target monitoring data is marked in chronological order and grouped according to the identified license plate numbers to obtain grouped marking results, and the marking results are compared with the target monitoring data in time and space to obtain the action sequence status, where the action status includes in-bounds, out-bounds and stable. The marking results and action status are stored in the cache database, and a monitoring time window is established to analyze behavioral characteristics. The monitoring vehicles that meet the preset alarm conditions are screened, and the alarmed vehicles are sent to the preset terminal device for processing through the terminal device. Through the spatiotemporal clustering model and incremental query mechanism, the dependence and cost of big data components are reduced, and the accuracy and real-time performance of the alarm are improved. This method effectively solves the shortcomings of traditional technologies such as high deployment costs and limitations in localized computing efficiency, significantly reducing deployment costs and improving localized computing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0017] Figure 1 Schematic diagram of a flow chart of a method for monitoring local vehicles in an embodiment of the present application; Figure 2 is a structural diagram of a device for monitoring local vehicles in an embodiment of the present application; Figure 3 Schematic diagram of the structure of the electronic device in the embodiment of the present application.

[0018] Reference numerals: Electronic device 9600, central processing unit 9100, memory 9140, communication module 9110, input unit 9120, audio processor 9130, display 9160, power supply 9170, buffer memory 9141, application / function storage unit 9142, data storage unit 9143, driver program storage unit 9144, antenna 9111, speaker 9131, microphone 9132. DETAILED DESCRIPTION

[0019] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0020] The acquisition, storage, use, and processing of data in this application's technical solution comply with relevant national laws and regulations.

[0021] In order to effectively solve the shortcomings of traditional technologies such as high deployment costs and limitations in localized computing efficiency, significantly reduce deployment costs, and improve localized computing efficiency, this application provides an embodiment of a method for monitoring local vehicles, see Figure 1 The method for monitoring local vehicles specifically includes the following contents: Step S101: receiving a query instruction, querying initial monitoring data that meets preset configuration conditions based on the query instruction, inputting the initial monitoring data into a trained spatiotemporal clustering model, and extracting target spatial clusters that meet a preset density threshold through the trained spatiotemporal clustering model.

[0022] The preset configuration conditions include a parking range and / or preset information data within the parking range.

[0023] Optionally, this embodiment receives a query instruction, wherein the query instruction is used to trigger a data retrieval operation, and determines the initial monitoring data that meets the preset configuration conditions based on the received query instruction, wherein the preset configuration conditions include but are not limited to the parking range and preset information data within the parking range, and may also include the parking range and preset data information within the parking range.

[0024] Furthermore, when the query instruction is used to query a specified monitored vehicle, the preset configuration condition corresponding to the initial monitoring data can be set to the parking range to determine the monitored vehicle, and the preset data information corresponding to the monitored vehicle can be obtained from the second preset query cycle, without the need to monitor within the parking range.

[0025] Correspondingly, when the query instruction is used to query the monitored vehicles within the parking range, the monitored vehicles within the parking range and their corresponding preset data information are determined in each preset query cycle.

[0026] Furthermore, the efficiency of querying initial monitoring data can be improved through the use of distributed databases and indexes. Among them, the distributed database can store large amounts of monitoring data in a dispersed manner across multiple server nodes, enabling parallel query and storage of data, improving query speed and storage capacity. The index can establish an index structure for parking ranges and preset information data. For example, a spatial index can quickly locate data that meets the parking range, and can quickly locate the initial monitoring data that meets the preset configuration conditions in a larger data set.

[0027] In addition, the acquired initial monitoring data is input into the trained spatiotemporal clustering model, wherein the spatiotemporal clustering model can perform a comprehensive analysis based on the spatial and temporal characteristics of the data and extract target spatial clusters that meet the preset density threshold. During the training process, the spatiotemporal clustering model can learn the aggregation and distribution patterns of vehicles at different times and locations, and identify cluster structures in which vehicles are relatively dense in space and continuous in time.

[0028] The target spatial cluster refers to a set of data points that meet a preset density threshold and are identified by a clustering algorithm, and appears as a high-density area in the spatial dimension.

[0029] Furthermore, the spatiotemporal clustering model can better determine the target spatial clusters that meet the preset density threshold through an architecture that combines recurrent neural networks and convolutional neural networks in deep learning.

[0030] This embodiment realizes the determination of initial monitoring data and the extraction of target spatial clusters that meet a preset density threshold based on a trained spatiotemporal clustering model. It can enhance the accuracy of identifying vehicle behavior patterns, reduce operation and maintenance costs, and achieve the ability to process large amounts of data in a lightweight and efficient manner.

[0031] Step S102: Determine the centroid coordinates and diffusion speed of each target spatial cluster. When the diffusion speed is greater than a preset movement threshold, generate a virtual monitoring vehicle including a centroid trajectory corresponding to the centroid coordinates, and determine the object corresponding to the virtual monitoring vehicle as the monitoring vehicle.

[0032] Optionally, this embodiment determines the centroid coordinates and diffusion speed of each extracted target space cluster, wherein the centroid coordinates can reflect the concentrated position of vehicles in the target space cluster, and the diffusion speed is used to represent the speed at which vehicles diffuse outward.

[0033] When the diffusion speed is greater than the preset movement threshold, the vehicles in the target space cluster as a whole show an obvious movement trend, and a virtual monitoring vehicle corresponding to the center of mass coordinates is generated. The center of mass coordinates corresponding to the virtual monitoring vehicle are continuously recorded to form a center of mass trajectory, and the physical object (i.e., object) mapped by the virtual monitoring vehicle is marked as a monitoring vehicle.

[0034] Among them, the target spatial cluster refers to a collection of points or objects with similar positions in spatial data, which is used to characterize the group behavior of vehicles in the monitoring scene. The center of mass trajectory provides a motion model for the virtual monitoring vehicle to simulate the movement pattern of the cluster and enhance the monitoring accuracy.

[0035] Furthermore, a vehicle can be regarded as a target space cluster, so as to more precisely identify and determine the monitored vehicle.

[0036] This embodiment realizes the determination of monitored vehicles through centroid coordinates and diffusion speed, improves the recognition accuracy of abnormal moving targets, and improves the computing efficiency in scenarios with large amounts of data.

[0037] Step S103: performing an incremental secondary query on the real-time monitoring data corresponding to the monitored vehicle according to the preset query cycle and the preset configuration conditions, and obtaining target monitoring data that meets the preset configuration conditions.

[0038] Optionally, this embodiment performs an incremental secondary query on the real-time monitoring data corresponding to the monitored vehicle according to a preset query cycle and preset configuration conditions. The incremental secondary query can be performed through a message queue and real-time stream processing. The message queue is used to buffer and orderly transmit the query request and the returned real-time monitoring data so that the query request can be sent to the data source in a timely and stable manner. The real-time stream processing method can perform real-time processing and analysis on the returned real-time monitoring data to obtain the target monitoring data that meets the preset configuration conditions.

[0039] The preset query period refers to a pre-set time interval or period, which is used to regulate the execution frequency of query operations and provide a time benchmark for secondary queries to ensure the periodicity and efficiency of data retrieval and avoid resource waste.

[0040] Among them, incremental secondary query refers to retrieving new or changed data since the previous query, rather than the entire data set. It can efficiently obtain updated data, reduce computing load and network overhead, and improve query performance.

[0041] Furthermore, the preset configuration condition may be the vehicles within the parking range in the initial monitoring data.

[0042] This embodiment implements incremental secondary query of real-time monitoring data through preset query cycles and preset configuration conditions, reduces data transmission volume and processing time, improves response speed, avoids big data components, and improves maintainability and scalability.

[0043] Step S104: Mark the target monitoring data in chronological order and group them according to the identified license plate numbers to obtain grouped marking results, compare the marking results with the target monitoring data in temporal and spatial correlation to obtain an action state sequence corresponding to the action state of the monitored vehicle.

[0044] Among them, the action states include in-bounds, out-bounds and stable.

[0045] Optionally, this embodiment marks the target monitoring data that meets the preset configuration conditions obtained through the secondary query in chronological order, and at the same time identifies the license plate number corresponding to the monitored vehicle, and groups the marked target monitoring data according to the identified license plate number. For example, the target monitoring data generated by the same vehicle at different time points can be grouped together, so that the behavior process of each vehicle can be fully tracked.

[0046] In addition, the grouped marking results are compared with the target monitoring data in time and space, that is, the state changes of the vehicle in different time and space are compared, so as to determine the action state sequence corresponding to the action state of the monitored vehicle, wherein the action state includes entry, exit and stability. Entry means that the vehicle enters the parking range, exit means that the vehicle leaves the parking range, and stability means that the vehicle is consistent with the parking range obtained in the last detection, that is, if the last detection shows that the vehicle is in the parking range and the current detection shows that the vehicle is still in the parking range, the vehicle's action state is determined to be stable. Similarly, if the last detection shows that the vehicle is outside the parking range and the current detection shows that the vehicle is still outside the parking range, the vehicle's action state is determined to be stable.

[0047] Among them, grouping is used to classify the target monitoring data in a specified way and organize it into multiple subsets, and to classify the marked target monitoring data according to the license plate number to form a structured data set.

[0048] Furthermore, in the process of marking the target monitoring data, it can be marked by a timestamp to more accurately reflect the time series of the vehicle's behavior.

[0049] This embodiment enhances the spatiotemporal continuity of vehicle behavior analysis, improves the accuracy of action recognition, and reduces resource consumption and operation and maintenance complexity.

[0050] Step S105: The marking results and the action status corresponding to the marking results are stored in a preset cache database, a monitoring time window is established for the monitored vehicle, the behavioral characteristics of the monitored vehicle are determined based on the marking results and the action status within the monitoring time window, and the monitored vehicles that meet the preset alarm conditions are screened based on the behavioral characteristics to obtain the corresponding alarm vehicles, and the alarm vehicles are sent to the preset terminal device so that the alarm vehicles can be processed through the terminal device.

[0051] Optionally, this embodiment stores the marking results and the action status corresponding to the marking results through a preset cache database, wherein the cache database can be a Redis database, which has the characteristics of fast reading and writing and temporary storage, and can store and quickly retrieve monitoring data and analysis results in a timely manner.

[0052] At the same time, a monitoring time window is established for the monitored vehicle. Within the monitoring time window, the marking results and action status are comprehensively considered, and the behavioral characteristics of the monitored vehicle are determined through data statistics and machine learning algorithms, such as the movement frequency, residence time distribution, and driving trajectory complexity of the monitored vehicle.

[0053] In addition, based on behavioral characteristics, monitored vehicles that meet the preset alarm conditions are screened, alarm vehicles are obtained, and the information of the alarm vehicles is sent to the preset terminal device so that relevant personnel can handle the alarm vehicles in time through the terminal device.

[0054] This embodiment achieves enhanced real-time and accuracy in monitoring vehicles, avoids the complexity and cost of deploying and maintaining a large-scale distributed computing framework, reduces resource consumption, and at the same time improves the computing efficiency and processing power of behavioral feature analysis of large amounts of monitoring data through localized algorithms and caching mechanisms.

[0055] This embodiment achieves efficient, accurate, and real-time monitoring and management of monitored vehicles by receiving query instructions and accurately querying initial monitoring data, extracting target spatial clusters through a spatiotemporal clustering model, and determining the center of mass coordinates and diffusion speed. It then performs real-time incremental secondary queries on the monitored vehicles, accurately marks groups and analyzes their action states, stores data in a cache database, and determines behavioral characteristics for alarm processing, thereby improving monitoring efficiency, reducing the workload and cost of manual monitoring, reducing deployment costs, and improving localized computing efficiency.

[0056] In some embodiments, an incremental secondary query is performed on the real-time monitoring data corresponding to the monitored vehicle according to a preset query period and preset configuration conditions, including: Receive the initial query cycle and preset buffer time, and obtain the previous query time; If there is a previous query time, the sum of the initial query period and the preset buffer time is determined as the preset query period, and the previous query time is used as the starting query time to perform an incremental secondary query on the real-time monitoring data corresponding to the time period of the preset query period; When there is no previous query time, the initial query period is determined as the preset query period, the starting query time is determined based on the query instruction, and an incremental secondary query is performed on the real-time monitoring data corresponding to the time period of the preset query period; When it is detected that the transmission delay of the real-time monitoring data exceeds the preset buffer time, a compensation query is performed on the real-time monitoring data with the transmission delay, and the preset buffer time is adjusted.

[0057] Optionally, this embodiment receives an initial query cycle and a preset buffer time, wherein the initial query cycle and the preset buffer time can be obtained according to user or system settings. For example, the initial query cycle can be set to 5 minutes, and the preset buffer time can be set to 2 minutes. At the same time, the preset buffer time can also be dynamically adjusted according to the real-time monitoring data processed and received.

[0058] At the same time, the previous query time is obtained to determine the current query time point.

[0059] In addition, when there is a previous query time, the sum of the initial query cycle and the preset buffer time is determined as the preset query cycle, that is, the previous query time is used as the starting query time point, and from this time point on, the real-time monitoring data within a time period of the preset query cycle is incrementally queried for a second time.

[0060] In addition, when there is no previous query time, the time point when the query instruction is received is determined as the starting query time point, and the initial query cycle is determined as the preset query cycle. The real-time monitoring data within the time period starting from the starting query time point and the length of the preset query cycle is incrementally queried for a second time.

[0061] In addition, during the query process, the transmission delay of the real-time monitoring data is detected. When it is detected that the transmission delay of the real-time detection data exceeds the preset buffer time, the compensation query mechanism is activated, that is, a supplementary query is performed on the delayed real-time monitoring data to ensure the integrity of the real-time monitoring data. At the same time, the preset buffer time is dynamically adjusted according to the delay time to adapt to changes in network conditions and optimize query performance.

[0062] Furthermore, the preset buffer time can be dynamically adjusted based on historical delay data and network conditions processed by a machine learning algorithm, wherein the machine learning algorithm can be regression analysis or reinforcement learning to adjust the query strategy in advance to minimize the impact of data loss and query delay.

[0063] This embodiment realizes stable operation in complex network environments by dynamically adjusting preset buffer time and compensatory query, reducing real-time monitoring data loss and monitoring loopholes, and improving the reliability of real-time monitoring data. By optimizing the preset query cycle, it can reduce unnecessary data processing and storage overhead, improve query performance and resource utilization efficiency, and reduce operating costs.

[0064] In some embodiments, the labeling result is compared with the target monitoring data in a temporal and spatial manner to obtain an action state sequence corresponding to the action state of the monitored vehicle, including: Assign a target name to each monitored vehicle and construct a spatiotemporal state matrix with the target name as the primary key. The row dimension of the spatiotemporal state matrix represents the time series, and the column dimension of the spatiotemporal state matrix includes the location coordinates of the monitored vehicles and the regional topological relationship. Determine the similarity between consecutive time slices in the current spatiotemporal state matrix, quantify the intensity of changes in the vehicle's action state in consecutive time slices based on the similarity metric, and insert an abnormal marker in the action state sequence when the intensity of the change exceeds a preset change threshold, where each time slice represents a row in the spatiotemporal state matrix.

[0065] Optionally, in this embodiment, a target name is assigned to each monitored vehicle, wherein the target name is a unique identifier, that is, it has uniqueness to ensure that different monitored vehicles can be distinguished.

[0066] Using the assigned target name as the primary key, a spatiotemporal state matrix is constructed, where the rows of the spatiotemporal state matrix are used to represent time series and can be arranged according to specified time intervals to record the state information of the monitored vehicle at different time points. The columns of the spatiotemporal state matrix are used to represent the location coordinates and regional topological relationships of the monitored vehicle, where the location coordinates can be represented by the longitude and latitude of the monitored vehicle, and the regional topological relationship can be the spatial relationship between the monitored vehicle and other surrounding vehicles or fixed landmarks.

[0067] In addition, the similarity between the multiple dimensional data of the current time slice and multiple adjacent time slices can be determined by similarity, wherein each time slice (i.e., each row in the spatiotemporal state matrix) represents the spatiotemporal state of the monitored vehicle at a certain moment, and the multiple dimensional data may include but are not limited to position coordinates and regional topological relationships, wherein the similarity can be determined according to multiple methods, such as Euclidean distance, cosine similarity or Hamming distance.

[0068] In addition, based on the similarity, the intensity of the change in the action state of the monitored vehicle between consecutive time slices can be quantified. The quantitative index of the change intensity can be defined according to the level of similarity. For example, the lower the similarity, the greater the change in the action state and the higher the change intensity.

[0069] In addition, when the change intensity exceeds the preset change threshold, an abnormal mark is inserted in the action state sequence, where the abnormal mark indicates that the action state of the monitored vehicle has changed significantly in the corresponding time slice, which may be the sudden acceleration, sudden braking, lane change or other unusual behavior of the monitored vehicle. The preset change threshold can be set based on historical data and expert experience to distinguish between normal behavior and abnormal behavior.

[0070] Furthermore, the spatiotemporal state matrix can be stored in a sparse matrix storage or compressed storage format, which can reduce storage space usage.

[0071] Furthermore, the preset change threshold can be dynamically adjusted based on the adaptive threshold adjustment algorithm of machine learning to automatically learn normal behavior patterns based on historical data, and the preset change threshold can be adjusted in real time according to changes in data distribution to meet the needs of anomaly detection in different scenarios.

[0072] This embodiment realizes the accurate tracking and recording of the state changes of each monitored vehicle in different time and space by constructing a spatiotemporal state matrix, which can more accurately capture the state changes of the monitored vehicles. At the same time, the introduction of a multi-dimensional similarity measurement algorithm and feature normalization processing can more reasonably evaluate the similarity and change intensity of the monitored vehicle states, avoid misjudgment due to data scale differences or nonlinear changes, reduce omissions and false alarms, and improve overall performance, scalability and stability.

[0073] In some embodiments, storing the marking results and the action states corresponding to the marking results in a preset cache database includes: The timestamp field in the marking result is determined as the primary index key, and the marking result is stored in a preset cache database in a columnar storage structure based on the primary index key; The marking results corresponding to hot data that exceeds the preset validity period are stored in distributed files, and the marking results corresponding to cold data that exceeds the preset validity period are stored in the memory database.

[0074] Optionally, this embodiment determines the timestamp field in the marking result as the primary index key, wherein the timestamp is unique and sequential in the real-time monitoring data and can accurately reflect the time characteristics of the real-time monitoring data.

[0075] At the same time, the marking results are stored in the preset cache database in a columnar storage structure based on the primary index key. Columnar storage organizes real-time monitoring data by columns, with each column serving as an independent storage unit. Compared with row-based storage, it is more efficient when querying certain specific fields. Especially for data analysis and aggregation operations, it can reduce the amount of data read and improve query performance.

[0076] In addition, the marking results corresponding to hot data that exceeds the preset expiration period are stored in distributed files. Hot data refers to data that is frequently accessed within a certain period of time. Distributed files (such as HDFS) can provide high-throughput data access, are suitable for storing large amounts of hot data, and have good scalability and fault tolerance.

[0077] At the same time, the marking results corresponding to cold data that exceeds the preset validity period are stored in the in-memory database. Cold data refers to data with low access frequency. In-memory databases (such as Redis) have extremely low latency and high read and write speeds, which are suitable for fast storage and retrieval of cold data, and can ensure that query requests can be responded to quickly even after the data becomes cold.

[0078] Furthermore, data compression algorithms can be applied to column storage for compression. The applied data compression algorithms can be, for example, dictionary encoding, run-length encoding, etc. The data types of the same column in column storage are the same and similar, and the compression effect is significant, which can reduce storage space occupancy and improve storage efficiency.

[0079] This embodiment achieves rapid storage and query of monitoring data, timely response to query requests, improved monitoring efficiency and real-time performance, enhanced data storage and management capabilities, and provides a foundation for long-term stable operation and efficient monitoring.

[0080] In some embodiments, the monitored vehicles that meet the preset alarm conditions are screened based on the behavioral characteristics to obtain the corresponding alarm vehicles, including: For each monitored vehicle, based on the behavioral characteristics and initial monitoring data corresponding to the current preset query cycle, if it is determined that the duration of the currently monitored vehicle's stay in the current parking area exceeds the preset retention duration and the monitored vehicle's motion state is stable, the monitored vehicle will be determined as an alarm vehicle, and the corresponding stay start time of the monitored vehicle will be updated to the start query time of the next preset query cycle, where the stay start time is the time point when the monitored vehicle first appears; For each monitored vehicle, based on the behavioral characteristics and initial monitoring data corresponding to the current preset query cycle, it is determined that the current monitored vehicle's stay time in the current parking range exceeds the preset stay time and the monitored vehicle's action state is out of bounds, and the corresponding stay time of the monitored vehicle is reset.

[0081] Optionally, this embodiment analyzes each monitored vehicle in combination with the behavioral characteristics and initial monitoring data corresponding to the current preset query cycle. If the monitored vehicle stays in the current parking range for longer than the preset detention time and the vehicle's motion state is stable, the monitored vehicle is determined to be an alarm vehicle, indicating that the monitored vehicle may have experienced an abnormally long and stable parking period and requires attention. The preset detention time can be 120 minutes.

[0082] In addition, the corresponding stop start time of the monitored vehicle is updated to the start query time of the next preset query cycle. The stop start time is used to indicate the corresponding time point when the monitored vehicle first appears in the current parking range, and is used to continue tracking the movement status of the monitored vehicle in subsequent preset query cycles to ensure that its movement status can be continuously monitored.

[0083] Similarly, for each monitored vehicle, a judgment is made based on the behavioral characteristics corresponding to the current preset query cycle and the initial monitoring data. If the monitored vehicle stays in the current parking range for longer than the preset retention time, and the action state of the monitored vehicle is out of bounds, that is, the monitored vehicle leaves the parking range after exceeding the retention time, the corresponding retention time of the monitored vehicle is reset so that the retention time of the monitored vehicle in the next parking range can be recalculated in the subsequent monitoring process to avoid interference with subsequent judgments due to the previous long retention time.

[0084] Furthermore, the state transitions of monitored vehicles can be represented by a finite state machine model. Each monitored vehicle can be in the action states of entering the parking range, stabilizing, and going out of bounds. State transitions are triggered according to different events. Among them, events may include but are not limited to stay time timeout and action state changes, thereby more clearly defining the behavior patterns and preset alarm conditions of the monitored vehicles and simplifying the implementation and maintenance of the alarm logic.

[0085] Furthermore, the atomicity and consistency of the update operations of the stay start time and stay duration can be ensured through the database transaction mechanism. That is, during the update process, either all relevant data are successfully updated or not updated, thereby avoiding data inconsistency problems caused by failures or concurrent operations.

[0086] This embodiment realizes determining whether a monitored vehicle is an alarm vehicle through monitoring data, improves the real-time and accuracy of the monitored vehicle detention alarm, and reduces the dependency cost of big data components.

[0087] In some embodiments, the monitored vehicles that meet the preset alarm conditions are screened based on the behavioral characteristics to obtain the corresponding alarm vehicles, including: For each monitored vehicle, based on the behavioral characteristics corresponding to the current preset query period and the initial monitoring data, determine the corresponding action state of the current monitored vehicle in the current parking range as the number of out-of-bounds and the number of in-bounds; The number of out-of-bounds and in-bounds is determined as one network touch count. If the number of network touch counts within a preset network touch cycle is not less than the preset network touch count, the monitored vehicle is determined as an alarm vehicle. When it is detected that the current preset network contact cycle ends, the number of network contacts in the current preset network contact cycle is cleared, and the next preset network contact cycle is started.

[0088] Optionally, this embodiment performs behavioral feature analysis on each monitored vehicle within each preset query cycle, and determines the corresponding action state of the current monitored vehicle within the current parking range as the number of out-of-bounds and the number of in-bounds, based on the behavioral features corresponding to the current preset query cycle and the initial monitoring data. That is, the action state sequence of the monitored vehicle is scanned, and whenever the monitored vehicle is detected to leave the current parking range, the number of out-of-bounds is increased by 1 based on the current record, and whenever the monitored vehicle re-enters the current parking range, the number of in-bounds is increased by 1 based on the current record, so as to reflect the frequency of the monitored vehicle's activities near the current parking range.

[0089] In addition, one out-of-bounds count and one in-bounds count are determined as one net touch count, wherein the net touch count is used to measure the frequency of the monitored vehicle entering and exiting the current parking range within a preset net touch cycle.

[0090] If the number of network touches is not less than the preset number of network touches within the preset network touch cycle, it indicates that the monitored vehicle frequently enters and exits the parking area within the preset network touch cycle, and may have abnormal behavior, such as wandering, probing, etc. In this case, the monitored vehicle will be identified as an alarm vehicle.

[0091] In addition, when it is detected that the current preset network contact cycle has ended, the number of network contacts within the current preset network contact cycle will be cleared to prepare for the next preset network contact cycle, so that the statistical independence of each network contact cycle can continuously and accurately monitor the behavior of the monitored vehicle and avoid the interference of historical data on the judgment of the current cycle.

[0092] Furthermore, a cycle management module can be set up to track and manage the start and end times of preset contact cycles. This cycle management module can switch between cycles based on a timer or time wheel algorithm. Furthermore, a dynamic adjustment mechanism can be introduced to automatically adjust the thresholds for the preset contact cycles and the number of contact attempts based on changes in the actual monitoring scenario and vehicle behavior. For example, a machine learning algorithm can analyze historical data to predict normal vehicle behavior patterns in different time periods, thereby appropriately adjusting alarm parameters to suit different traffic conditions and monitoring needs.

[0093] This embodiment implements the screening of monitored vehicles that meet preset alarm conditions based on behavioral characteristics, realizes the accurate and timely identification of abnormal behaviors of monitored vehicles, improves the efficiency and accuracy of monitoring, enhances flexibility and adaptability, and enables it to better meet the traffic management and safety monitoring needs in different scenarios.

[0094] In order to effectively solve the shortcomings of traditional technologies such as high deployment costs and limitations in localized computing efficiency, significantly reduce deployment costs, and improve localized computing efficiency, this application provides an embodiment of a device for monitoring local vehicles for realizing the monitoring of all or part of the local vehicles. Figure 2 The device for monitoring local vehicles specifically includes the following contents: A first processing module 10 is configured to receive a query instruction, query initial monitoring data that meets preset configuration conditions based on the query instruction, input the initial monitoring data into a trained spatiotemporal clustering model, and extract target spatial clusters that meet a preset density threshold using the trained spatiotemporal clustering model, wherein the preset configuration conditions include a parking range and / or preset information data within the parking range; The second processing module 20 is configured to determine the centroid coordinates and diffusion speed of each target spatial cluster. When the diffusion speed is greater than a preset movement threshold, a virtual monitoring vehicle including a centroid trajectory corresponding to the centroid coordinates is generated, and the object corresponding to the virtual monitoring vehicle is determined as the monitoring vehicle. The third processing module 30 is used to perform an incremental secondary query on the real-time monitoring data corresponding to the monitored vehicle according to a preset query cycle and preset configuration conditions to obtain target monitoring data that meets the preset configuration conditions; The fourth processing module 40 is configured to mark the target monitoring data in chronological order and group them according to the recognized license plate numbers to obtain grouped marking results, compare the marking results with the target monitoring data in temporal and spatial correlation, and obtain an action state sequence corresponding to the action state of the monitored vehicle, where the action state includes in-bounds, out-of-bounds, and stable. The fifth processing module 50 is used to store the marking results and the action status corresponding to the marking results through a preset cache database, establish a monitoring time window for the monitored vehicle, determine the behavioral characteristics of the monitored vehicle based on the marking results and the action status within the monitoring time window, and screen the monitored vehicles that meet the preset alarm conditions based on the behavioral characteristics, obtain the corresponding alarm vehicle, and send the alarm vehicle to a preset terminal device so that the alarm vehicle can be processed through the terminal device.

[0095] As can be seen from the above description, the device for monitoring local vehicles provided in the embodiment of the present application can obtain initial monitoring data that meets preset configuration conditions by receiving query instructions, and input the initial monitoring data into the spatiotemporal clustering model to extract target spatial clusters that meet preset density thresholds, wherein the preset configuration conditions include parking ranges and / or preset information data within the parking ranges, determine the centroid coordinates and diffusion speed of each target spatial cluster, and when the diffusion speed is greater than the preset movement threshold, generate a virtual monitoring vehicle including a centroid trajectory corresponding to the centroid coordinates, determine the object corresponding to the virtual monitoring vehicle as the monitoring vehicle, and obtain the target monitoring vehicle through a periodic incremental query corresponding to a preset period. Data, target monitoring data is marked in chronological order and grouped according to the identified license plate numbers to obtain grouped marking results, and the marking results are compared with the target monitoring data in time and space to obtain the action sequence status, where the action status includes in-bounds, out-bounds and stable. The marking results and action status are stored in the cache database, and a monitoring time window is established to analyze behavioral characteristics. The monitoring vehicles that meet the preset alarm conditions are screened, and the alarmed vehicles are sent to the preset terminal device for processing through the terminal device. Through the spatiotemporal clustering model and incremental query mechanism, the dependence and cost of big data components are reduced, and the accuracy and real-time performance of the alarm are improved. This method effectively solves the shortcomings of traditional technologies such as high deployment costs and limitations in localized computing efficiency, significantly reducing deployment costs and improving localized computing efficiency.

[0096] From a hardware perspective, in order to effectively address the shortcomings of traditional technologies such as high deployment costs and limitations in localized computing efficiency, significantly reduce deployment costs, and improve localized computing efficiency, this application provides an embodiment of an electronic device for implementing all or part of the method for monitoring local vehicles. The electronic device specifically includes the following: A processor, memory, a communications interface, and a bus; wherein the processor, memory, and communications interface communicate with each other via the bus; the communications interface is used to transmit information between the device for monitoring local vehicles and related devices such as core business systems, user terminals, and related databases; the logic controller can be a desktop computer, a tablet computer, a mobile terminal, etc., but this embodiment is not limited thereto. In this embodiment, the logic controller can be implemented with reference to the embodiments of the method for monitoring local vehicles and the embodiments of the device for monitoring local vehicles in the embodiments, the contents of which are incorporated herein and any repetitions are omitted.

[0097] It is understandable that the user terminal may include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device may include smart glasses, a smart watch, a smart bracelet, etc.

[0098] In practical applications, portions of the method for monitoring local vehicles may be executed on the electronic device as described above, or all operations may be performed on the client device. The specific method may be selected based on the processing capabilities of the client device and the limitations of the user's usage scenario. This application does not impose any restrictions on this. If all operations are performed on the client device, the client device may also include a processor.

[0099] The aforementioned client device may include a communication module (i.e., a communication unit) capable of establishing a communication connection with a remote server to facilitate data transmission with the server. The server may include a server at the task scheduling center or, in other implementation scenarios, a server on an intermediate platform, such as a server on a third-party server platform that is communicatively linked to the task scheduling center server. The server may comprise a single computer device, a server cluster consisting of multiple servers, or a distributed server configuration.

[0100] Figure 3 Schematic block diagram of the system structure of the electronic device 9600 according to an embodiment of the present application. Figure 3 As shown, the electronic device 9600 may include a central processing unit 9100 and a memory 9140; the memory 9140 is coupled to the central processing unit 9100. It is worth noting that the Figure 3 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0101] In one embodiment, the method and functionality for monitoring local vehicles may be integrated into the central processing unit 9100. The central processing unit 9100 may be configured to perform the following control: Step S101: Receive a query instruction, query initial monitoring data that meets preset configuration conditions based on the query instruction, input the initial monitoring data into a trained spatiotemporal clustering model, and extract target spatial clusters that meet a preset density threshold using the trained spatiotemporal clustering model. The preset configuration conditions include parking areas and / or preset information data within the parking areas. Step S102: Determine the centroid coordinates and diffusion speed of each target spatial cluster. When the diffusion speed is greater than a preset movement threshold, generate a virtual monitoring vehicle including a centroid trajectory corresponding to the centroid coordinates, and determine the object corresponding to the virtual monitoring vehicle as the monitoring vehicle. Step S103: performing an incremental secondary query on the real-time monitoring data corresponding to the monitored vehicle according to a preset query cycle and preset configuration conditions to obtain target monitoring data that meets the preset configuration conditions; Step S104: Marking the target monitoring data in chronological order and grouping them according to the recognized license plate numbers to obtain grouped marking results. The marking results are then compared with the target monitoring data for temporal and spatial correlation to obtain an action state sequence corresponding to the action state of the monitored vehicle. The action states include in-bounds, out-of-bounds, and stable. Step S105: The marking results and the action status corresponding to the marking results are stored in a preset cache database, a monitoring time window is established for the monitored vehicle, the behavioral characteristics of the monitored vehicle are determined based on the marking results and the action status within the monitoring time window, and the monitored vehicles that meet the preset alarm conditions are screened based on the behavioral characteristics to obtain the corresponding alarm vehicles, and the alarm vehicles are sent to the preset terminal device so that the alarm vehicles can be processed through the terminal device.

[0102] As can be seen from the above description, the electronic device provided in the embodiment of the present application obtains initial monitoring data that meets the preset configuration conditions by receiving a query instruction, inputs the initial monitoring data into the spatiotemporal clustering model to extract target spatial clusters that meet the preset density threshold, wherein the preset configuration conditions include the parking range and / or preset information data within the parking range, determines the centroid coordinates and diffusion speed of each target spatial cluster, and when the diffusion speed is greater than the preset movement threshold, generates a virtual monitoring vehicle including a centroid trajectory corresponding to the centroid coordinates, determines the object corresponding to the virtual monitoring vehicle as the monitoring vehicle, obtains the target monitoring data through a periodic incremental query corresponding to the preset period, and performs the following operations on the target spatial clusters: The target monitoring data is marked in chronological order and grouped according to the identified license plate numbers to obtain the grouped marking results. The marking results are then compared with the target monitoring data in a spatiotemporal manner to obtain the action sequence status, where the action status includes in-bounds, out-of-bounds, and stable. The marking results and action status are stored in a cache database, and a monitoring time window is established to analyze behavioral characteristics. Monitoring vehicles that meet the preset alarm conditions are screened, and the alarmed vehicles are sent to the preset terminal device for processing by the terminal device. Through the spatiotemporal clustering model and incremental query mechanism, the dependence and cost of big data components are reduced, and the accuracy and real-time performance of alarms are improved. This method effectively addresses the shortcomings of traditional technologies such as high deployment costs and limitations in localized computing efficiency, significantly reducing deployment costs and improving localized computing efficiency.

[0103] In another embodiment, the device for monitoring local vehicles can be configured separately from the central processor 9100. For example, the device for monitoring local vehicles can be configured as a chip connected to the central processor 9100, and the method function of monitoring local vehicles can be implemented through the control of the central processor.

[0104] like Figure 3 As shown, the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily have to include Figure 3 In addition, the electronic device 9600 may also include all components shown in Figure 3 For components not shown, reference may be made to the prior art.

[0105] like Figure 3 As shown, the central processing unit 9100 is sometimes also referred to as a controller or operation control, and may include a microprocessor or other processor device and / or logic device. The central processing unit 9100 receives input and controls the operation of various components of the electronic device 9600.

[0106] Memory 9140 can be, for example, one or more of a cache, flash memory, hard drive, removable media, volatile memory, non-volatile memory, or other suitable devices. It can store the aforementioned failure-related information and also store programs that execute the relevant information. The CPU 9100 can execute the programs stored in memory 9140 to implement information storage or processing.

[0107] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 may be, for example, a keypad or touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used to display objects such as images and text. The display may be, for example, an LCD display, but is not limited thereto.

[0108] The memory 9140 may be a solid-state memory, such as a read-only memory (ROM), random access memory (RAM), or SIM card. Alternatively, it may be a memory that retains information even when power is off, can be selectively erased, and is capable of storing additional data. Examples of such memory are sometimes referred to as EPROMs. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142 for storing application programs and function programs, or processes used by the central processing unit 9100 to execute operations of the electronic device 9600.

[0109] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, images, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various driver programs for communication functions of the electronic device and / or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0110] The communication module 9110 is a transmitter / receiver that transmits and receives signals via the antenna 9111. The communication module 9110 (transmitter / receiver) is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal.

[0111] Based on different communication technologies, multiple communication modules 9110 may be provided in the same electronic device, such as cellular network modules, Bluetooth modules, and / or wireless local area network modules. The communication module 9110 (transmitter / receiver) is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130, providing audio output via the speaker 9131 and receiving audio input from the microphone 9132, thereby implementing common telecommunication functions. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. Furthermore, the audio processor 9130 is coupled to the central processing unit 9100, enabling local recording via the microphone 9132 and playback of stored audio via the speaker 9131.

[0112] Embodiments of the present application also provide a computer-readable storage medium capable of implementing all steps of the method for monitoring a local vehicle in the above-mentioned embodiment, where the execution subject is a server or a client. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the computer program implements all steps of the method for monitoring a local vehicle in the above-mentioned embodiment, where the execution subject is a server or a client. For example, when the processor executes the computer program, the following steps are implemented: Step S101: Receive a query instruction, query initial monitoring data that meets preset configuration conditions based on the query instruction, input the initial monitoring data into a trained spatiotemporal clustering model, and extract target spatial clusters that meet a preset density threshold using the trained spatiotemporal clustering model. The preset configuration conditions include parking areas and / or preset information data within the parking areas. Step S102: Determine the centroid coordinates and diffusion speed of each target spatial cluster. When the diffusion speed is greater than a preset movement threshold, generate a virtual monitoring vehicle including a centroid trajectory corresponding to the centroid coordinates, and determine the object corresponding to the virtual monitoring vehicle as the monitoring vehicle. Step S103: performing an incremental secondary query on the real-time monitoring data corresponding to the monitored vehicle according to a preset query cycle and preset configuration conditions to obtain target monitoring data that meets the preset configuration conditions; Step S104: Marking the target monitoring data in chronological order and grouping them according to the recognized license plate numbers to obtain grouped marking results. The marking results are then compared with the target monitoring data for temporal and spatial correlation to obtain an action state sequence corresponding to the action state of the monitored vehicle. The action states include in-bounds, out-of-bounds, and stable. Step S105: The marking results and the action status corresponding to the marking results are stored in a preset cache database, a monitoring time window is established for the monitored vehicle, the behavioral characteristics of the monitored vehicle are determined based on the marking results and the action status within the monitoring time window, and the monitored vehicles that meet the preset alarm conditions are screened based on the behavioral characteristics to obtain the corresponding alarm vehicles, and the alarm vehicles are sent to the preset terminal device so that the alarm vehicles can be processed through the terminal device.

[0113] As can be seen from the above description, the computer-readable storage medium provided in the embodiment of the present application obtains initial monitoring data that meets the preset configuration conditions by receiving a query instruction, inputs the initial monitoring data into the spatiotemporal clustering model to extract target spatial clusters that meet the preset density threshold, wherein the preset configuration conditions include the parking range and / or preset information data within the parking range, determines the centroid coordinates and diffusion speed of each target spatial cluster, and when the diffusion speed is greater than the preset movement threshold, generates a virtual monitoring vehicle including a centroid trajectory corresponding to the centroid coordinates, determines the object corresponding to the virtual monitoring vehicle as the monitoring vehicle, and obtains the target monitoring data through a periodic incremental query corresponding to the preset period. According to the data, the target monitoring data is marked in chronological order and grouped according to the identified license plate numbers to obtain the grouped marking results. The marking results are compared with the target monitoring data in time and space to obtain the action sequence status, where the action status includes in-bounds, out-bounds and stable. The marking results and action status are stored in the cache database, and a monitoring time window is established to analyze the behavioral characteristics. The monitoring vehicles that meet the preset alarm conditions are screened, and the alarmed vehicles are sent to the preset terminal device to process the alarmed vehicles through the terminal device. Through the spatiotemporal clustering model and incremental query mechanism, the dependence and cost of big data components are reduced, and the accuracy and real-time performance of the alarm are improved. This method effectively solves the shortcomings of traditional technologies such as high deployment costs and limitations in localized computing efficiency, significantly reducing deployment costs and improving localized computing efficiency.

[0114] Embodiments of the present application also provide a computer program product capable of implementing all steps of the method for monitoring a local vehicle in the above-mentioned embodiment, where the execution subject is a server or a client. When the computer program / instructions are executed by a processor, the steps of the method for monitoring a local vehicle are implemented. For example, the computer program / instructions implement the following steps: Step S101: Receive a query instruction, query initial monitoring data that meets preset configuration conditions based on the query instruction, input the initial monitoring data into a trained spatiotemporal clustering model, and extract target spatial clusters that meet a preset density threshold using the trained spatiotemporal clustering model. The preset configuration conditions include parking areas and / or preset information data within the parking areas. Step S102: Determine the centroid coordinates and diffusion speed of each target spatial cluster. When the diffusion speed is greater than a preset movement threshold, generate a virtual monitoring vehicle including a centroid trajectory corresponding to the centroid coordinates, and determine the object corresponding to the virtual monitoring vehicle as the monitoring vehicle. Step S103: performing an incremental secondary query on the real-time monitoring data corresponding to the monitored vehicle according to a preset query cycle and preset configuration conditions to obtain target monitoring data that meets the preset configuration conditions; Step S104: Marking the target monitoring data in chronological order and grouping them according to the recognized license plate numbers to obtain grouped marking results. The marking results are then compared with the target monitoring data for temporal and spatial correlation to obtain an action state sequence corresponding to the action state of the monitored vehicle. The action states include in-bounds, out-of-bounds, and stable. Step S105: The marking results and the action status corresponding to the marking results are stored in a preset cache database, a monitoring time window is established for the monitored vehicle, the behavioral characteristics of the monitored vehicle are determined based on the marking results and the action status within the monitoring time window, and the monitored vehicles that meet the preset alarm conditions are screened based on the behavioral characteristics to obtain the corresponding alarm vehicles, and the alarm vehicles are sent to the preset terminal device so that the alarm vehicles can be processed through the terminal device.

[0115] As can be seen from the above description, the computer program product provided in the embodiment of the present application obtains initial monitoring data that meets preset configuration conditions by receiving a query instruction, inputs the initial monitoring data into a spatiotemporal clustering model to extract target spatial clusters that meet a preset density threshold, wherein the preset configuration conditions include a parking range and / or preset information data within the parking range, determines the centroid coordinates and diffusion speed of each target spatial cluster, and when the diffusion speed is greater than a preset movement threshold, generates a virtual monitoring vehicle including a centroid trajectory corresponding to the centroid coordinates, determines the object corresponding to the virtual monitoring vehicle as a monitoring vehicle, and obtains the target monitoring data through a periodic incremental query corresponding to a preset period. , the target monitoring data is marked in chronological order and grouped according to the identified license plate numbers to obtain the grouped marking results. The marking results are compared with the target monitoring data in time and space to obtain the action sequence status, where the action status includes in-bounds, out-bounds, and stable. The marking results and action status are stored in the cache database, and a monitoring time window is established to analyze the behavioral characteristics. The monitored vehicles that meet the preset alarm conditions are screened, and the alarmed vehicles are sent to the preset terminal device for processing through the terminal device. Through the spatiotemporal clustering model and incremental query mechanism, the dependence and cost of big data components are reduced, and the accuracy and real-time performance of the alarm are improved. This method effectively solves the shortcomings of traditional technologies such as high deployment costs and limitations in localized computing efficiency, significantly reducing deployment costs and improving localized computing efficiency.

[0116] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, apparatuses, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0117] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (apparatus), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0118] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0119] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0120] Specific embodiments are used in the present invention to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core ideas. At the same time, for those skilled in the art, according to the ideas of the present invention, there may be changes in the specific implementation methods and application scopes. In summary, the contents of this specification should not be understood as limiting the present invention.

Claims

1. A method for monitoring local vehicles, characterized in that: The method comprises: receiving a query instruction, querying initial monitoring data that meets preset configuration conditions based on the query instruction, inputting the initial monitoring data into a trained spatiotemporal clustering model, and extracting target spatial clusters that meet a preset density threshold using the trained spatiotemporal clustering model, wherein the preset configuration conditions include a parking range and / or preset information data within the parking range; Determining the centroid coordinates and diffusion speed of each target spatial cluster; when the diffusion speed is greater than a preset movement threshold, generating a virtual monitoring vehicle including a centroid trajectory corresponding to the centroid coordinates; and determining the object corresponding to the virtual monitoring vehicle as a monitoring vehicle; According to a preset query cycle and the preset configuration conditions, an incremental secondary query is performed on the real-time monitoring data corresponding to the monitored vehicle to obtain target monitoring data that meets the preset configuration conditions; Marking the target monitoring data in chronological order and grouping them according to the identified license plate numbers to obtain grouped marking results, performing spatiotemporal correlation comparison between the marking results and the target monitoring data to obtain an action state sequence corresponding to the action state of the monitored vehicle, wherein the action state includes in-bounds, out-of-bounds, and stable; The marking results and the action states corresponding to the marking results are stored in a preset cache database, a monitoring time window is established for the monitored vehicle, the behavioral characteristics of the monitored vehicle are determined based on the marking results and the action states within the monitoring time window, and the monitored vehicles that meet the preset alarm conditions are screened based on the behavioral characteristics to obtain the corresponding alarm vehicles, and the alarm vehicles are sent to a preset terminal device so that the alarm vehicles can be processed by the terminal device.

2. The method according to claim 1, characterized in that The performing of an incremental secondary query on the real-time monitoring data corresponding to the monitored vehicle according to the preset query period and the preset configuration condition includes: Receive the initial query cycle and preset buffer time, and obtain the previous query time; When the previous query time exists, the sum of the initial query period and the preset buffer time is determined as the preset query period, and the previous query time is used as the starting query time to perform an incremental secondary query on the real-time monitoring data corresponding to the time period of the preset query period; When the previous query time does not exist, the initial query period is determined as the preset query period, a starting query time is determined based on the query instruction, and an incremental secondary query is performed on the real-time monitoring data corresponding to the time period of the preset query period; When it is detected that the transmission delay of the real-time monitoring data exceeds the preset buffer time, a compensation query is performed on the real-time monitoring data with the transmission delay, and the preset buffer time is adjusted.

3. The method according to claim 1, characterized in that The step of performing a temporal and spatial correlation comparison between the labeling result and the target monitoring data to obtain an action state sequence corresponding to the action state of the monitored vehicle includes: Assigning a target name to each of the monitored vehicles, and constructing a spatiotemporal state matrix with the target name as a primary key, wherein the row dimension of the spatiotemporal state matrix represents a time series, and the column dimension of the spatiotemporal state matrix includes the location coordinates and regional topological relationships of the monitored vehicles; Determine the similarity between consecutive time slices in the current spatiotemporal state matrix, quantify the intensity of the change of the action state of the monitored vehicle in the consecutive time slices based on the similarity, and insert an abnormal mark in the action state sequence when the change intensity exceeds a preset change threshold, wherein each time slice represents a row in the spatiotemporal state matrix.

4. The method according to claim 1, wherein The storing the marking result and the action state corresponding to the marking result in a preset cache database includes: Determine the timestamp field in the marking result as a primary index key, and store the marking result in the preset cache database in a column storage structure based on the primary index key; The marking results corresponding to the hot data exceeding the preset validity period are stored in a distributed file, and the marking results corresponding to the cold data exceeding the preset validity period are stored in a memory database.

5. The method according to claim 1, wherein The step of screening the monitored vehicles that meet the preset alarm conditions based on the behavioral characteristics to obtain the corresponding alarm vehicles includes: For each monitored vehicle, based on the behavioral characteristics corresponding to the current preset query cycle and the initial monitoring data, if it is determined that the duration of stay of the current monitored vehicle in the current parking range exceeds the preset retention duration and the action state of the monitored vehicle is stable, the monitored vehicle is determined as an alarm vehicle, and the corresponding stay start time of the monitored vehicle is updated to the start query time of the next preset query cycle, wherein the stay start time is the time point when the monitored vehicle first appears; For each monitored vehicle, based on the behavioral characteristics corresponding to the current preset query cycle and the initial monitoring data, it is determined that the stay time of the current monitored vehicle in the current parking range exceeds the preset stay time and the action state of the monitored vehicle is out of bounds, and the corresponding stay time of the monitored vehicle is reset.

6. The method according to claim 1, characterized in that The step of screening the monitored vehicles that meet the preset alarm conditions based on the behavioral characteristics to obtain the corresponding alarm vehicles includes: For each monitored vehicle, based on the behavioral characteristics corresponding to the current preset query period and the initial monitoring data, determining the action state corresponding to the current monitored vehicle within the current parking range as the out-of-bounds number of times and the in-bounds number of times; Determine one of the out-of-bounds times and one of the in-bounds times as one network touch count, and determine the monitored vehicle as an alarm vehicle if the network touch count is not less than a preset network touch count within a preset network touch cycle; When it is detected that the current preset network contact cycle ends, the number of network contacts in the current preset network contact cycle is cleared, and the next preset network contact cycle is started.

7. A device for monitoring local vehicles, characterized in that: The device comprises: a first processing module configured to receive a query instruction, query initial monitoring data that meets preset configuration conditions based on the query instruction, input the initial monitoring data into a trained spatiotemporal clustering model, and extract target spatial clusters that meet a preset density threshold using the trained spatiotemporal clustering model, wherein the preset configuration conditions include a parking range and / or preset information data within the parking range; a second processing module, configured to determine the centroid coordinates and diffusion speed of each target spatial cluster, and when the diffusion speed is greater than a preset movement threshold, generate a virtual monitoring vehicle including a centroid trajectory corresponding to the centroid coordinates, and determine the object corresponding to the virtual monitoring vehicle as the monitoring vehicle; A third processing module is configured to perform an incremental secondary query on the real-time monitoring data corresponding to the monitored vehicle according to a preset query cycle and the preset configuration conditions, to obtain target monitoring data that meets the preset configuration conditions; a fourth processing module, configured to mark the target monitoring data in chronological order and group them according to the identified license plate numbers to obtain grouped marking results, perform spatiotemporal correlation comparison between the marking results and the target monitoring data, and obtain an action state sequence corresponding to the action state of the monitored vehicle, wherein the action state includes in-bounds, out-of-bounds, and stable; The fifth processing module is used to store the marking results and the action status corresponding to the marking results through a preset cache database, establish a monitoring time window for the monitored vehicle, determine the behavioral characteristics of the monitored vehicle based on the marking results and the action status within the monitoring time window, and screen the monitored vehicles that meet the preset alarm conditions based on the behavioral characteristics, obtain the corresponding alarm vehicle, and send the alarm vehicle to a preset terminal device so that the alarm vehicle can be processed by the terminal device.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the method for monitoring a local vehicle according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for monitoring a local vehicle according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method for monitoring a local vehicle according to any one of claims 1 to 6 are implemented.

Images