Data processing method, device, equipment and medium for dealing with emergencies

Abstract

The present invention provides a data processing method, device, equipment and readable medium for dealing with emergencies, which captures different data sources, processes the data of the different data sources to form a comprehensive data pool; screens the data in the comprehensive data pool to generate a data set; analyzes and calculates the data set to generate a specific character portrait; scores the specific character portrait using an evaluation model; and determines whether the specific character portrait needs to be processed. The present invention breaks the information barriers between the different data sources, forms a comprehensive data acquisition system, realizes real-time and accurate capture of crowd activity trajectories, and greatly improves the coverage and response speed of public security management.

Description

[Technical field]

[0001] The present invention relates to the field of data processing technology, and in particular to a data processing method, device, equipment and medium for coping with emergencies. [Background technology]

[0002] In recent years, with the advancement of the "Skynet" operation, the deployment of photographic equipment to accurately locate and capture people in public places has alleviated the probability of illegal and criminal activities to a certain extent, increased the predictability of emergencies, and also increased social stability to a certain extent.

[0003] For example, in the prior art with patent number CN202310889616.8 and patent name A method, device, equipment and medium for detecting abnormal behavior based on video stream, the prior art obtains the video stream to be detected in real time, and performs image acquisition on the video stream to be detected according to a preset image acquisition method to obtain the picture stream to be detected; if it is determined that the picture stream to be detected meets the classification model input conditions, the picture stream to be detected is input into the pre-trained picture stream classification model to obtain the picture stream category and the confidence corresponding to the picture stream category; if according to the picture stream category and the confidence, the type of the video stream to be detected is determined to be abnormal behavior, and a real-time alarm processing operation is performed. This solves the technical problem that fighting and brawling cannot be accurately identified and alarmed in large-scale video surveillance scenarios, improves the accuracy of detecting abnormal behavior, and can better maintain social security.

[0004] For example, in the patent number CN202011464516.3, the patent name is a vehicle identification and tracking method, which discloses a vehicle identification and tracking method in the field of vehicle identification and tracking technology, including the following steps:

[0005] S1. Obtain image information of the vehicle to be tracked, and determine the target vehicle to be identified and tracked; S2. Find several videos of the target vehicle captured by the highway camera, and then train the region divider, convolution kernel, classifier, and regressor through the videos to obtain trained network parameters for target identification of the target specific vehicle; The present invention provides a scientific method for identifying and tracking criminal vehicles using existing traffic monitoring system resources, provides technical support for the police to pursue criminals, can improve the success rate of solving cases and maintain social order to a certain extent, and automatically identifies targets by machine, which greatly reduces the time and cost of manually finding all relevant videos, and prevents omissions and unclear situations during manual search.

[0006] However, with the increase in the number of camera devices and the number of departments using cameras, the interaction between cameras in various departments has become extremely difficult. Data islands are obvious between systems. For example, data resources such as train ticket purchase, bus ticket purchase, airplane ticket purchase, checkpoint vehicle passing records, travel check-in information, portrait capture, and video surveillance are scattered and unconnected, and cannot be effectively and centrally analyzed and processed in real time. In addition, when using various social data for target recognition, the existing methods for dealing with emergencies also cause the video surveillance warning to have low accuracy and poor real-time performance due to the similar reasons that data resources such as video surveillance are scattered and unconnected, resulting in poor tracking of crowd dynamics. In addition, due to the similar reasons that data resources such as video surveillance are scattered and unconnected, there may be a lag in the prediction of crowd behavior patterns and potential risk assessment, making it difficult to achieve real-time warning and rapid response to emergencies. [Summary of the invention]

[0007] In view of the above technical problems in the prior art, the present invention provides a data processing method for dealing with emergencies, comprising:

[0008] Capture different data sources, and process the data from the different data sources to form a comprehensive data pool;

[0009] Screening the data in the comprehensive data pool to generate a data set;

[0010] Analyzing and calculating the data set to generate a portrait of a specific person;

[0011] Using the evaluation model to score the specific character portrait;

[0012] When the score of the specific character portrait is greater than or equal to the preset warning value, the specific character portrait is intervened;

[0013] When the score of the specific character portrait is less than the preset warning value, there is no need to process the specific character portrait.

[0014] It should be noted that the process of using the evaluation model to score the specific character portrait includes:

[0015] Defining the different data sources includes: data source A, data source B, data source C, data source D and data source E;

[0016] The score of the data source A is calculated as follows:

[0017] (0-100 points)*weight a%, one time is recorded as A0, the number of times the data source A appears is recorded as A1, and the score A2=A0*A1;

[0018] The score calculation method of the data source B is:

[0019] (0-100 points)*weight b%, one time is recorded as B0, the number of times the data source B appears is recorded as B1, and the score B2=B0*B1;

[0020] The score of the data source C is calculated as follows:

[0021] (0-100 points)*weight c%, where one occurrence is recorded as C0, the number of occurrences of the data source C is recorded as C1, and the score C2=C0*C1;

[0022] The score of the data source D is calculated as follows:

[0023] (0-100 points)*weight d%, one time is recorded as D0, the number of times the data source D appears is recorded as D1, and the score D2=D0*D1;

[0024] The score calculation method of the data source E is:

[0025] (0-100 points)*weight e%, one time is recorded as E0, the number of times the data source E appears is recorded as E1, and the score E2=E0*E1;

[0026] The meanings of a, b, c, d, and e are defined as the proportions of the data source A, the data source B, the data source C, the data source D, and the data source E in the different data sources, that is, the sum of a, b, c, d, and e must be 100;

[0027] The A0, B0, C0, D0, and E0 are user-defined constants;

[0028] Define the values ​​of A0, B0, C0, D0, E0, A2, B2, C2, D2, and E2 to be between 0 and 100 respectively;

[0029] The score of the specific character portrait is defined as X;

[0030] Said X=A2*a%+B2*b%+C2*c%+D2*d%+E2*e%.

[0031] It should be noted that the steps of capturing data from different data sources and processing the data from different data sources to form a comprehensive data pool include:

[0032] By reading data, the data from different data sources are captured;

[0033] Arrange the data from different data sources so that the data from different data sources form data in a unified format;

[0034] The data in the unified format are processed to form the comprehensive data pool.

[0035] It should be noted that the step of screening the data in the comprehensive data pool to generate a data set includes:

[0036] The data in the comprehensive data pool is subjected to missing value filling, and / or outlier detection, and / or data type conversion operations to generate the data set.

[0037] It should be noted that the step of analyzing and calculating the data set to generate a specific character portrait includes:

[0038] Performing in-depth analysis on the data set to calculate the relationship between different data in the data set;

[0039] constructing an individual behavior pattern based on the relationship between different data in the data set;

[0040] The specific character portrait is generated according to the behavior pattern.

[0041] The present invention also provides a data processing device for dealing with emergencies, comprising:

[0042] A capture module, used to capture different data sources and process the data from the different data sources to form a comprehensive data pool;

[0043] A screening module, used for screening the data in the comprehensive data pool to generate a data set;

[0044] An analysis module, used to analyze and calculate the data set to generate a portrait of a specific person;

[0045] An evaluation module is used to score the specific character portrait through the evaluation model;

[0046] When the score of the specific character portrait is greater than or equal to the preset warning value, the specific character portrait is intervened;

[0047] When the score of the specific character portrait is less than the preset warning value, there is no need to process the specific character portrait.

[0048] It should be noted that the capture module includes:

[0049] A reading unit, used to obtain the data from the different data sources by means of data reading;

[0050] A sorting unit, used to sort the data from different data sources so that the data from different data sources form data in a unified format;

[0051] The comprehensive data pool unit is used to process the data in the unified format to form the comprehensive data pool.

[0052] It should be noted that the screening module includes:

[0053] A conversion operation unit, used for performing missing value filling, outlier detection, and / or data type conversion operations on the data in the comprehensive data pool to generate the data set;

[0054] Wherein, the data set includes crowd behavior characteristic information.

[0055] It should be noted that the analysis module includes:

[0056] A deep analysis unit, used to perform deep analysis on the data set and calculate the relationship between different data in the data set;

[0057] A construction unit, used for constructing an individual behavior pattern according to the relationship between different data in the data set;

[0058] A generating unit is used to generate the specific character portrait according to the behavior pattern.

[0059] The present invention also 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 data processing method for dealing with emergencies when executing the computer program.

[0060] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to implement the data processing method for dealing with emergencies when executed by a processor.

[0061] The technical solution of the embodiment of the present invention captures different data sources and processes the data of the different data sources to form a comprehensive data pool; screens the data in the comprehensive data pool to generate a data set; analyzes and calculates the data set to generate a specific character portrait; scores the specific character portrait using an evaluation model; when the score of the specific character portrait is greater than or equal to a preset warning value, the specific character portrait is intervened; when the score of the specific character portrait is less than the preset warning value, the specific character portrait does not need to be processed. The present invention can process the data of the different data sources, break the information barriers between the different data sources, and form a comprehensive data acquisition system. Compared with the problem of single data source and delayed update in the prior art, the present invention realizes real-time and accurate capture of crowd activity trajectories, greatly improving the coverage and response speed of public security management. Using advanced big data mining technology and deep learning algorithms, the present invention can automatically identify key behavior patterns from complex social activities, and conduct high-precision risk assessment and warning based on them. Based on the real-time data-driven decision support system provided by the present invention, the public security management department can quickly deploy police resources according to the real-time crowd dynamic information, reduce ineffective patrols and blind spot monitoring, effectively improve the efficiency of the use of police resources, and ensure rapid response and proper handling of emergencies.

[0062] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Brief Description of the Drawings

[0063] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work, among which:

[0064] Figure 1 It is a flow chart of the data processing method for dealing with emergencies disclosed in the present invention;

[0065] Figure 2 Yes Figure 1 The flowchart shown is for capturing data from different data sources and processing the data from different data sources to form a comprehensive data pool;

[0066] Figure 3 Yes Figure 1The process diagram shown is for screening the data in the comprehensive data pool to generate a data set;

[0067] Figure 4 Yes Figure 1 The schematic diagram of the process of analyzing and calculating the data set to generate a portrait of a specific person;

[0068] Figure 5 It is a structural schematic diagram of a data processing device for dealing with emergencies disclosed in the present invention;

[0069] Figure 6 Yes Figure 5 The structural diagram of the grabbing module shown;

[0070] Figure 7 Yes Figure 5 The structural schematic diagram of the screening module shown;

[0071] Figure 8 Yes Figure 5 The structural diagram of the analysis module shown;

[0072] Fig. 9 It is a structural schematic diagram of an electronic device disclosed by the present invention. [Specific implementation method]

[0073] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0074] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0075] In addition, in the present invention, descriptions such as "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0076] In the present invention, unless otherwise clearly specified and limited, the terms "connection", "fixation", etc. should be understood in a broad sense. For example, "fixation" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0077] In addition, the technical solutions between the various embodiments of the present invention can be combined with each other, but it must be based on the fact that ordinary technicians in the field can implement it. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0078] like Figure 1 As shown, Figure 1 The present invention discloses a data processing method for dealing with emergencies, which includes the following steps:

[0079] S1. Capture different data sources, and process the data from the different data sources to form a comprehensive data pool;

[0080] S2. Screening the data in the comprehensive data pool to generate a data set;

[0081] S3, analyzing and calculating the data set to generate a portrait of a specific person;

[0082] S4, scoring the specific character portrait using the evaluation model;

[0083] When the score of the specific character portrait is greater than or equal to the preset warning value, the specific character portrait is intervened;

[0084] When the score of the specific character portrait is less than the preset warning value, there is no need to process the specific character portrait.

[0085] Please also see Figure 2 , Figure 2 Yes Figure 1 The schematic diagram of the process of S1 shown;

[0086] It should be noted that the process of using the evaluation model to score the specific character portrait includes:

[0087] Defining the different data sources includes: data source A, data source B, data source C, data source D and data source E;

[0088] The score of the data source A is calculated as follows:

[0089] (0-100 points)*weight a%, one time is recorded as A0, the number of times the data source A appears is recorded as A1, and the score A2=A0*A1;

[0090] The score calculation method of the data source B is:

[0091] (0-100 points)*weight b%, one time is recorded as B0, the number of times the data source B appears is recorded as B1, and the score B2=B0*B1;

[0092] The score of the data source C is calculated as follows:

[0093] (0-100 points)*weight c%, where one occurrence is recorded as C0, the number of occurrences of the data source C is recorded as C1, and the score C2=C0*C1;

[0094] The score of the data source D is calculated as follows:

[0095] (0-100 points)*weight d%, one time is recorded as D0, the number of times the data source D appears is recorded as D1, and the score D2=D0*D1;

[0096] The score calculation method of the data source E is:

[0097] (0-100 points)*weight e%, one time is recorded as E0, the number of times the data source E appears is recorded as E1, and the score E2=E0*E1;

[0098] The meanings of a, b, c, d, and e are defined as the proportions of the data source A, the data source B, the data source C, the data source D, and the data source E in the different data sources, that is, the sum of a, b, c, d, and e must be 100;

[0099] The A0, B0, C0, D0, and E0 are user-defined constants;

[0100] Define the values ​​of A0, B0, C0, D0, E0, A2, B2, C2, D2, and E2 to be between 0 and 100 respectively;

[0101] The score of the specific character portrait is defined as X;

[0102] Said X=A2*a%+B2*b%+C2*c%+D2*d%+E2*e%.

[0103] The step S1 comprises:

[0104] S11, capturing data from different data sources by means of data reading;

[0105] S12, arranging the data from different data sources so that the data from different data sources form data in a unified format;

[0106] S13. Process the data in the unified format to form the comprehensive data pool.

[0107] Furthermore, in S11, the data can be read by calling an API interface, using a web crawler, or reading data from a real-time hardware sensor.

[0108] It should be noted that the API interface call, web crawler technology, and real-time hardware sensor data reading methods can be used simultaneously, or several of them can be used simultaneously. The API interface call, web crawler technology, and real-time hardware sensor methods are not considered to be a limitation on the steps of capturing different data sources in the claims. As long as the method of capturing data from different data sources is implemented, it is within the protection scope of the present invention.

[0109] It should be noted that API (Application Programming Interface) is a way for software systems to access and exchange information with each other. It defines the way to call another application (or service) from one application. The API interface allows developers to interact only through the API interface without having to understand the specific implementation details within the program, thereby improving the reusability and interoperability of the software. Generally speaking, the API interface includes some functions, protocols, data structures, and tool classes. Open APIs can help companies provide better products and services, while also increasing the scope of use and interactivity of services. They are an indispensable part of modern products and services.

[0110] It should be noted that, in S11, the data reading process needs to be processed as follows:

[0111] First, data encryption processing, that is, encrypting data through encryption algorithms to ensure the security of data during transmission. Common encryption algorithms include symmetric encryption (such as AES) and asymmetric encryption (such as RSA). Ensure that the encryption process can avoid affecting the company's business while meeting the security requirements of data transmission.

[0112] Among them, the Advanced Encryption Standard (AES) in cryptography, also known as Rijndael encryption, is a block encryption standard adopted by the US federal government.

[0113] Among them, the RSA algorithm is an asymmetric encryption algorithm. Unlike the symmetric encryption algorithm, the RSA algorithm has two different keys, one is the public key and the other is the private key.

[0114] Second, identity authentication and permission control, that is, using identity authentication technology to confirm the identity of the transmission node and ensure the security of the transmission node. For example, by using digital certificates or role-based access control (RBAC) and other methods. Perform integrity detection on the data that has completed encrypted transmission to confirm whether the integrity of the data has been destroyed after the transmission is completed. If the integrity is destroyed, the system should have the ability to tolerate data faults or recover data.

[0115] Third, data transmission channel security protection, that is, focus on the security between proxy server and terminal, proxy server and Internet, and proxy server and proxy server. Strengthen the security of key transmission channels, such as using encryption protocols such as HTTPS instead of HTTP to prevent data from being stolen or tampered with during transmission. Conduct security audits and monitoring regularly to detect and prevent potential security threats.

[0116] Among them, HTTPS (full name: Hypertext Transfer Protocol Secure) is an HTTP channel with security as its goal. It ensures the security of the transmission process through transmission encryption and identity authentication based on HTTP.

[0117] Among them, HTTP, or Hypertext Transfer Protocol (HTTP), is a simple request-response protocol that usually runs on top of TCP. It specifies what kind of messages a client may send to a server and what kind of responses it may get.

[0118] Fourth, data storage security, that is, select secure storage devices and media, and configure and manage them securely. Ensure that only authorized personnel can access stored data. Conduct data backup and recovery drills regularly to deal with possible unexpected situations. Transmit the collected vehicle data to the cloud server through the mobile communication network for subsequent processing and analysis. The transmission method can use wireless communication technologies such as 4G and 5G, or near-field communication technologies such as WIFI and Bluetooth.

[0119] Furthermore, the different data sources in S11 may be various social data such as travel activities, accommodation activities, video surveillance, etc.

[0120] It should be noted that the process of processing the data in the unified format in S13 includes performing preliminary deduplication and standardization processing on the data in the unified format.

[0121] It should be noted that, in step S1 of the present invention, the data from different data sources are processed to break down the information barriers between the different data sources, thereby forming a comprehensive data acquisition system.

[0122] Please also see Figure 3 and Figure 1 , Figure 3 Yes Figure 1 The process diagram of S2 shown in FIG. 1 is as follows; S2 includes:

[0123] S21. Perform missing value filling, and / or outlier detection, and / or data type conversion operations on the data in the comprehensive data pool to generate the data set.

[0124] It should be noted that the missing value filling, and / or outlier value detection, and / or data type conversion operations mean that any one of the missing value filling, outlier value detection, and data type conversion operations can be used, or several of them can be operated simultaneously.

[0125] Wherein, the missing value filling may be to record the missing information and then fill it with the average value;

[0126] The outlier detection may be to remove abnormal data; or to merge repeated information and retain only one record.

[0127] Please also see Figure 4 and Figure 1 , Figure 4 Yes Figure 1 The process diagram of S3 shown in FIG. 1 is as follows; S3 includes:

[0128] S31, performing in-depth analysis on the data set to calculate the relationship between different data in the data set;

[0129] S32, constructing an individual behavior pattern according to the relationship between different data in the data set;

[0130] S33: Generate the specific person portrait according to the behavior pattern.

[0131] Furthermore, the in-depth analysis process requires the use of machine learning language for in-depth analysis.

[0132] Furthermore, the behavior pattern in step S33 includes the behavior feature information of the specific person and the spatiotemporal trajectory of the specific person.

[0133] It should be noted that the specific meaning of the step of constructing an individual behavior pattern based on the relationship between different data in the data set is:

[0134] Different social data are deeply analyzed and calculated, the different social data are screened and analyzed, and individual behavior patterns are constructed, thereby generating the specific character portrait.

[0135] It should be noted that in step S31, in the step of performing in-depth analysis on the data set, a clustering algorithm may be used to divide individuals into different groups, such as K-Means.

[0136] Before performing K-Means clustering, principal component analysis (PCA) is used to reduce the dimensionality of individual features:

[0137] The principal components analysis (PCA) is a technique for simplifying a data set. It is a linear transformation. This transformation transforms the data into a new coordinate system so that the first largest variance of any data projection is on the first coordinate (called the first principal component), the second largest variance is on the second coordinate (the second principal component), and so on.

[0138] K-Means clustering is the task of dividing data into multiple groups, each of which is called a cluster. The goal of clustering is to divide the data so that the elements in each group are very similar, but the data in different groups are very different. In simple terms, it is called classification.

[0139] Furthermore, in the step of using principal component analysis (PCA) to reduce the dimension of individual features, the following operations can be performed:

[0140] (1) Standardized data: Standardize individual feature data to ensure that the mean of each feature is 0 and the variance is 1.

[0141] Furthermore, standardization includes standardizing each column feature, which specifically includes the following steps:

[0142] (2) Calculate the covariance matrix: Calculate the covariance matrix of the standardized data.

[0143] (3) Calculate eigenvalues ​​and eigenvectors: Decompose the covariance matrix into eigenvalues ​​and eigenvectors.

[0144] (4) Select principal components: Select the eigenvectors corresponding to the largest k eigenvalues ​​to form a transformation matrix.

[0145] (5) Select the first k largest eigenvalues ​​and their corresponding eigenvectors:

[0146] (6) Dimensionality reduction transformation: Multiply the original data by the transformation matrix to obtain the reduced-dimensional data.

[0147] Furthermore, principal component analysis (PCA) is used to reduce the dimension of individual features, and the reduced features are subjected to K-Means clustering, wherein the steps of K-Means clustering include:

[0148] Initialization, that is, select K=2 and randomly initialize the centroid.

[0149] Assignment, that is, calculate the distance from each data point to the centroid and assign it to the group corresponding to the nearest centroid.

[0150] Update the centroids, that is, calculate the new centroid of each group, which is the average of all data points in the group.

[0151] Iteration, i.e., repeating the steps of assigning and updating the centroid, until the centroid no longer changes or a predetermined number of iterations is reached.

[0152] It should be noted that step S3 in the present invention utilizes advanced big data mining technology and deep learning algorithms, and the present invention can automatically identify key behavior patterns from complex social activities, and conduct high-precision risk assessment and early warning based on them. Compared with the limitations of existing technologies in identifying complex behavior patterns, the present invention can more accurately predict potential risks, thereby achieving precise control of the crowd and improving the level of public safety protection.

[0153] It should be noted that S4 includes:

[0154] S41, tracking the specific person portrait in combination with the location and time information;

[0155] S42, after tracking the specific person portrait, using the evaluation model to perform real-time evaluation and scoring on the specific person portrait;

[0156] When the score of the specific character portrait is greater than or equal to the preset warning value, the specific character portrait is intervened;

[0157] When the score of the specific character portrait is less than the preset warning value, there is no need to process the specific character portrait.

[0158] It should be noted that the steps of intervention include:

[0159] S4211, locking the location information of the specific person portrait;

[0160] S4212, pushing the location information to the administrative department corresponding to the location information;

[0161] S4213. After receiving the location information, the administrative department intervenes in the portrait of the specific person.

[0162] It should be noted that the step of intervening in the portrait of the specific person may be to make people in the spatial area where the specific person is located appear, leave, or gather.

[0163] It should be noted that, in one embodiment of the present invention, the process of making the evaluation model may include:

[0164] First, the model building stage;

[0165] A Multilayer Perceptron (MLP) is used as an example model.

[0166] It should be noted that MLP is the abbreviation of Multi-Layer Perceptron, which is a machine learning model based on feedforward neural network. It usually consists of multiple fully connected layers, each of which contains multiple neurons, each of which is connected to all neurons in the previous layer, and performs nonlinear transformation on the input through activation function.

[0167] The input layer of the model needs to match the number of features of the ticket purchase data, and the number of nodes in the output layer needs to match the number of personnel behavior categories.

[0168] Second, the model training stage;

[0169] The model is trained using the prepared dataset. We use the cross entropy loss function for optimization of multi-classification problems.

[0170] Third, model evaluation stage;

[0171] Evaluate the model and see how it performs on the training and validation sets.

[0172] It should be noted that the step of using the evaluation model to score the specific character portrait includes:

[0173] Behavioral feature extraction: Perform the same feature extraction and preprocessing on the data as during model training to ensure that the data format is consistent with the model input.

[0174] Use pre-trained control models for real-time scoring; Use pre-trained control models to score data in real time; Obtain the probability distribution of each behavior to indicate the likelihood of different categories of personnel behavior.

[0175] Scoring threshold setting and warning triggering: Set the scoring threshold, and when the highest probability of ticket purchase behavior exceeds the threshold, the warning mechanism is triggered. According to the specific scenario and needs, the appropriate threshold can be set according to the model's accuracy, recall rate and other indicators.

[0176] Furthermore, the step of intervening in the specific character portrait may also include:

[0177] Process the list of specific character portraits to obtain their detailed information and location coordinates;

[0178] Assess the risk level of specific personas, generate and push reports;

[0179] Interventions and decisions are made based on the reports.

[0180] It should be noted that step S4 in the present invention is based on the real-time data-driven decision support system provided by the present invention. The public security management department can quickly deploy police resources according to the real-time crowd dynamic information, reduce ineffective patrols and blind spot monitoring, effectively improve the efficiency of the use of police resources, and ensure rapid response and proper handling of emergencies.

[0181] See also Figure 5 , Figure 5 It is a structural schematic diagram of a data processing device 10 for dealing with emergencies disclosed in the present invention.

[0182] The present invention also provides a data processing device 10 for dealing with emergencies, comprising:

[0183] The capture module 11 is used to capture different data sources and process the data from the different data sources to form a comprehensive data pool;

[0184] A screening module 12, used for screening the data in the comprehensive data pool to generate a data set;

[0185] An analysis module 13 is used to analyze and calculate the data set to generate a portrait of a specific person;

[0186] An evaluation module 14, used to score the specific character portrait through an evaluation model;

[0187] When the score of the specific character portrait is greater than or equal to the preset warning value, the specific character portrait is intervened;

[0188] When the score of the specific character portrait is less than the preset warning value, there is no need to process the specific character portrait.

[0189] See also Figure 6 , Figure 6 Yes Figure 5 The structural diagram of the grabbing module shown in FIG. 1 ; the grabbing module 11 comprises:

[0190] A reading unit 111 is used to obtain data from different data sources by reading data;

[0191] The arranging unit 112 is used to arrange the data from different data sources so that the data from different data sources form data in a unified format;

[0192] The integrated data pool unit 113 is used to process the data in the unified format to form the integrated data pool.

[0193] See also Figure 7 , Figure 7 Yes Figure 5 The structural diagram of the screening module shown in FIG. 1 , the screening module 12 comprises:

[0194] A conversion operation unit 121, configured to perform missing value filling, outlier detection, and / or data type conversion operations on the data in the comprehensive data pool to generate the data set;

[0195] Wherein, the data set includes crowd behavior characteristic information.

[0196] See also Figure 8 , Figure 8 Yes Figure 5 The structural diagram of the analysis module shown in FIG. 1 , the analysis module 13 includes:

[0197] A deep analysis unit 131, configured to perform deep analysis on the data set and calculate the relationship between different data in the data set;

[0198] A construction unit 132, configured to construct an individual behavior pattern according to the relationship between different data in the data set;

[0199] The generating unit 133 is used to generate the specific person portrait according to the behavior pattern.

[0200] like Fig. 9 As shown, the present invention also provides an electronic device 20, comprising: a memory 21; a processor 22; and a computer program stored in the memory and executable on the processor, wherein the processor implements the data processing method for dealing with emergencies when executing the computer program.

[0201] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to implement the data processing method for dealing with emergencies when executed by a processor.

[0202] Compared with the prior art, the technical solution of the embodiment of the present invention captures different data sources and processes the data of the different data sources to form a comprehensive data pool; screens the data in the comprehensive data pool to generate a data set; analyzes and calculates the data set to generate a specific character portrait; scores the specific character portrait using an evaluation model; when the score of the specific character portrait is greater than or equal to the preset warning value, the specific character portrait is intervened; when the score of the specific character portrait is less than the preset warning value, the specific character portrait does not need to be processed. The present invention can process the data of the different data sources, break the information barriers between the different data sources, and form a comprehensive data acquisition system. Compared with the problem of single data source and delayed update in the prior art, the present invention realizes real-time and accurate capture of crowd activity trajectories, greatly improving the coverage and response speed of public security management. Using advanced big data mining technology and deep learning algorithms, the present invention can automatically identify key behavior patterns from complex social activities, and conduct high-precision risk assessment and warning based on them. Based on the real-time data-driven decision support system provided by the present invention, the public security management department can quickly deploy police resources according to the real-time crowd dynamic information, reduce ineffective patrols and blind spot monitoring, effectively improve the efficiency of the use of police resources, and ensure rapid response and proper handling of emergencies.

[0203] Since the device, electronic device and readable medium in the present invention adopt the above method, their corresponding technical effects are not described in detail here.

[0204] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A data processing method for dealing with emergencies, characterized in that: include: Capture different data sources, and process the data from the different data sources to form a comprehensive data pool; Screening the data in the comprehensive data pool to generate a data set; Analyzing and calculating the data set to generate a portrait of a specific person; Using the evaluation model to score the specific character portrait; When the score of the specific character portrait is greater than or equal to the preset warning value, the specific character portrait is intervened; When the score of the specific character portrait is less than the preset warning value, there is no need to process the specific character portrait; The process of scoring the specific character portrait using the evaluation model includes: Defining the different data sources includes: data source A, data source B, data source C, data source D and data source E; The score of the data source A is calculated as follows: (0-100 points)*weight a%, one time is recorded as A0, the number of times the data source A appears is recorded as A1, and the score A2=A0*A1; The score calculation method of the data source B is: (0-100 points)*weight b%, one time is recorded as B0, the number of times the data source B appears is recorded as B1, and the score B2=B0*B1; The score of the data source C is calculated as follows: (0-100 points)*weight c%, where one occurrence is recorded as C0, the number of occurrences of the data source C is recorded as C1, and the score C2=C0*C1; The score of the data source D is calculated as follows: (0-100 points)*weight d%, one time is recorded as D0, the number of times the data source D appears is recorded as D1, and the score D2=D0*D1; The score of the data source E is calculated as follows: (0-100 points)*weight e%, one time is recorded as E0, the number of times the data source E appears is recorded as E1, and the score E2=E0*E1; The meanings of a, b, c, d, and e are defined as the proportions of the data source A, the data source B, the data source C, the data source D, and the data source E in the different data sources, that is, the sum of a, b, c, d, and e must be 100; The A0, B0, C0, D0, and E0 are user-defined constants; Define the values ​​of A0, B0, C0, D0, E0, A2, B2, C2, D2, and E2 to be between 0 and 100 respectively; The score of the specific character portrait is defined as X; Said X=A2*a%+B2*b%+C2*c%+D2*d%+E2*e%.

2. The data processing method for dealing with emergencies according to claim 1, characterized in that: The step of capturing data from different data sources and processing the data from different data sources to form a comprehensive data pool includes: By reading data, the data from different data sources are captured; Arrange the data from different data sources so that the data from different data sources form data in a unified format; The data in the unified format are processed to form the comprehensive data pool.

3. The data processing method for dealing with emergencies according to claim 1, characterized in that: The step of screening the data in the comprehensive data pool to generate a data set includes: The data in the comprehensive data pool is subjected to missing value filling, and / or outlier detection, and / or data type conversion operations to generate the data set.

4. The data processing method for dealing with emergencies according to claim 1, characterized in that: The step of analyzing and calculating the data set to generate a specific character portrait includes: Performing in-depth analysis on the data set to calculate the relationship between different data in the data set; constructing an individual behavior pattern based on the relationship between different data in the data set; The specific character portrait is generated according to the behavior pattern.

5. A data processing device for dealing with emergencies, characterized in that: include: A capture module, used to capture different data sources and process the data from the different data sources to form a comprehensive data pool; A screening module, used for screening the data in the comprehensive data pool to generate a data set; An analysis module, used to analyze and calculate the data set to generate a portrait of a specific person; An evaluation module is used to score the specific character portrait through the evaluation model; When the score of the specific character portrait is greater than or equal to the preset warning value, the specific character portrait is intervened; When the score of the specific character portrait is less than the preset warning value, there is no need to process the specific character portrait; The working process of the evaluation module includes: Defining the different data sources includes: data source A, data source B, data source C, data source D and data source E; The score of the data source A is calculated as follows: (0-100 points)*weight a%, one time is recorded as A0, the number of times the data source A appears is recorded as A1, and the score A2=A0*A1; The score calculation method of the data source B is: (0-100 points)*weight b%, one time is recorded as B0, the number of times the data source B appears is recorded as B1, and the score B2=B0*B1; The score of the data source C is calculated as follows: (0-100 points)*weight c%, where one occurrence is recorded as C0, the number of occurrences of the data source C is recorded as C1, and the score C2=C0*C1; The score of the data source D is calculated as follows: (0-100 points)*weight d%, one time is recorded as D0, the number of times the data source D appears is recorded as D1, and the score D2=D0*D1; The score of the data source E is calculated as follows: (0-100 points)*weight e%, one time is recorded as E0, the number of times the data source E appears is recorded as E1, and the score E2=E0*E1; The meanings of a, b, c, d, and e are defined as the proportions of the data source A, the data source B, the data source C, the data source D, and the data source E in the different data sources, that is, the sum of a, b, c, d, and e must be 100; The A0, B0, C0, D0, and E0 are user-defined constants; Define the values ​​of A0, B0, C0, D0, E0, A2, B2, C2, D2, and E2 to be between 0 and 100 respectively; The score of the specific character portrait is defined as X; Said X=A2*a%+B2*b%+C2*c%+D2*d%+E2*e%.

6. The data processing device for dealing with emergencies according to claim 5, characterized in that: The capture module comprises: A reading unit, used to obtain the data from the different data sources by means of data reading; A sorting unit, used to sort the data from different data sources so that the data from different data sources form data in a unified format; The comprehensive data pool unit is used to process the data in the unified format to form the comprehensive data pool.

7. The data processing device for dealing with emergencies according to claim 5, characterized in that: The screening module comprises: A conversion operation unit, used for performing missing value filling, outlier detection, and / or data type conversion operations on the data in the comprehensive data pool to generate the data set; Wherein, the data set includes crowd behavior characteristic information.

8. The data processing device for dealing with emergencies according to claim 5, characterized in that: The analysis module comprises: A deep analysis unit, used to perform deep analysis on the data set and calculate the relationship between different data in the data set; A construction unit, used for constructing an individual behavior pattern according to the relationship between different data in the data set; A generating unit is used to generate the specific character portrait according to the behavior pattern.

9. An electronic device, comprising: Memory; processor; and A computer program stored in the memory and executable on the processor, wherein the processor implements the data processing method for coping with emergencies as described in any one of claims 1 to 5 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the data processing method for coping with emergencies according to any one of claims 1 to 5 when executed.

Images