Intelligent monitoring system and method based on urban security and protection

Abstract

The invention discloses an intelligent monitoring system and method based on city security and protection, relates to the technical field of monitoring systems, and solves the problems that an existing smart city security and protection monitoring system adopts a globally unified security and protection monitoring method, security and protection monitoring is difficult to carry out in a targeted manner according to the characteristics of all regions in a city, and the security and protection efficiency is high. And the overall efficiency of urban security and protection is reduced. The data acquisition module is used for acquiring detection data of each management area; the data processing module is used for extracting a plurality of regional feature data and monitoring video data in the detection data; generating a region supervision score of the management region according to the plurality of pieces of monitoring video data; generating a region supervision level corresponding to the management region according to the region feature data; the security early warning module is used for generating a security early warning signal according to the regional supervision level and the corresponding regional supervision score; performing early warning according to the security early warning signal; and the accuracy of the security and protection intelligent monitoring system in each management area of the city is improved.

Description

Technical Field

[0001] This application belongs to the field of monitoring technology, relates to monitoring technology, and specifically is a smart monitoring system and method based on urban security. Background Art

[0002] Smart monitoring is used to monitor urban areas, objects or processes in real time; it can monitor the flow of people; and give early warnings before security crises occur, which can effectively improve the reliability and stability of smart monitoring.

[0003] The prior art (CN117522017B) discloses a smart city security monitoring system based on image recognition technology, which relates to the field of monitoring system technology. The system discloses an image recognition module, a security warning module, and a security preset module. By setting the security warning module, it can judge whether it is necessary to send security warning information to security personnel within the security warning range according to the change of security risks, and give early warnings before security crises occur, so that security personnel can dispose of security incidents in a timely and efficient manner. By setting the security preset module, the number of security personnel corresponding to the monitoring equipment within the security warning range can be adjusted. On the basis of reducing the probability of security accidents near the monitoring equipment, ensure that a reasonable number of security personnel are set near each monitoring equipment, and adjust the security personnel in real time according to the security situation.

[0004] The above-mentioned smart city security monitoring system based on image recognition technology realizes unified monitoring of the entire city by collecting video data captured by monitoring equipment; each current city consists of several regions, and the functions and personnel compositions of different regions are different, and the requirements for security are also different. It is difficult to adapt to the specific characteristics of each different region by using a globally unified security monitoring method; resulting in unreasonable distribution of security monitoring in the city; it may cause over - security in some regions and insufficient security in some regions; thus leading to a decline in the overall efficiency of urban security.

[0005] Therefore, there is an urgent need for a smart monitoring system and method based on urban security. Summary of the Invention

[0006] This application aims to solve at least one of the technical problems existing in the prior art; for this purpose, this application proposes an intelligent monitoring system and method based on urban security, which is used to solve the technical problem that the existing urban intelligent security monitoring system adopts a globally unified security monitoring method and it is difficult to conduct targeted security monitoring according to the characteristics of each area in the city, resulting in a decline in the overall efficiency of urban security. This application solves the above problems by obtaining the detection data of each management area, extracting several regional feature data and monitoring video data from the detection data, generating a regional supervision score of the management area according to the several monitoring video data; and generating a corresponding regional supervision level of the management area according to the regional feature data, and then generating a security warning signal according to the regional supervision level and its corresponding regional supervision score.

[0007] To achieve the above object, the first aspect of this application provides an intelligent monitoring system and method based on urban security, including: a data acquisition module, a data processing module, a security warning module, and a database;

[0008] The data acquisition module: obtains the detection data of each management area through the corresponding data acquisition device;

[0009] The data processing module: extracts several regional feature data and monitoring video data from the detection data; generates a regional supervision score of the management area according to the several monitoring video data; and generates a corresponding regional supervision level of the management area according to the regional feature data;

[0010] The security warning module: generates a security warning signal according to the regional supervision level and its corresponding regional supervision score; issues a warning according to the security warning signal.

[0011] Further, generating the regional supervision score of the management area according to the several monitoring video data includes:

[0012] Obtaining several monitoring video data, extracting frames from the monitoring video data according to the set frame extraction rule to obtain several monitoring images; inputting the monitoring images into an image evaluation model to obtain the image recognition score corresponding to the monitoring images; generating a regional supervision score according to the image recognition scores corresponding to each monitoring image.

[0013] Further, the image evaluation model is trained by an artificial intelligence model and includes:

[0014] Obtain a number of monitoring images and their corresponding image recognition scores through a database; the image recognition score is an evaluation of the monitoring image by an expert based on the clarity and color presentation ability of the monitoring image. The higher the clarity of the monitoring image and the better the color presentation ability, it indicates that the quality of the video data collected by the corresponding monitoring device is higher, the more accurate the result of its recognition, and the larger the corresponding image recognition score is set; integrate the monitoring images and their corresponding image recognition scores into a number of training data and test data;

[0015] Use the training data to train the artificial intelligence model, and use the test data to test the trained artificial intelligence model; specifically, input the monitoring images in the test data into the trained artificial intelligence model to output the image recognition score. Determine whether the difference between the image recognition score and the image recognition score recorded in the test data is within the acceptable range; if so, it means that this group of test data passes the test, and continue to test the next group of test data; if not, the relevant parameters of the artificial intelligence model need to be adjusted, and continue to use this group of test data for testing; until a set proportion of the test data passes the test; finally, obtain an image evaluation model with the input being the monitoring image and its corresponding image recognition score and the output being the image recognition score, where the artificial intelligence model is a convolutional neural network model.

[0016] Furthermore, generating the area supervision score according to the image recognition scores corresponding to each monitoring image includes:

[0017] Obtain the image recognition scores of each monitoring image and mark them as TX ij ; where, i is the number of the monitoring device corresponding to the monitoring image, and j is the frame number corresponding to the monitoring image;

[0018] Through the formula Calculate to obtain the area supervision score TA corresponding to the management area; where, j = 1, 2,..., J, J represents the total number of frames corresponding to the monitoring image; α i Represents the weight coefficient of the corresponding monitoring image; i = 1, 2,..., F, F represents the total number of monitoring devices corresponding to the management area; Among them, f(i) represents an increasing function; obtain the average value of the image recognition scores of each monitoring image through the formula to obtain the area supervision score corresponding to the management area; the closer the monitoring image corresponding to the image recognition score is to the size of the monitoring area, the greater its weight value; in addition, in this embodiment, f(i) = log(1 + i + c), where c represents a constant and is greater than 0; f(i) can also be specifically set to other increasing functions about i according to experience, and the weight coefficients α i Are different.

[0019] This application obtains the area supervision score corresponding to the management area by averaging the image recognition scores of each monitoring image; the closer the monitoring image corresponding to the image recognition score is to the current moment, the greater its weight value; thus improving the accuracy by adding security equipment subsequently.

[0020] Further, generating the area supervision level corresponding to the management area according to the area feature data includes:

[0021] Extracting the personnel data and building data in the area feature data;

[0022] Generating a personnel influence coefficient according to the number of permanent residents, the number of floating people, the personnel density, and the educational level in the personnel data;

[0023] Generating a building influence coefficient according to the building life data and living environment data in the building data;

[0024] Generating the area supervision level according to the personnel influence coefficient and the building influence coefficient.

[0025] Further, generating the personnel influence coefficient according to the number of permanent residents, the number of floating people, the personnel density, and the educational level distribution includes:

[0026] According to the number of permanent residents, the number of floating people, and the personnel density in the management area in the personnel data; the educational level distribution in the management area specifically includes: D1 is the number of people receiving compulsory education in the management area, D2 is the number of people receiving higher education in the management area, and D3 is the number of people not receiving education in the management area; D z is the number of people corresponding to the educational level distribution numbered z; z = 1, 2, 3;

[0027] Through the formula Calculate the personnel influence coefficient PA within the management area; a z represents the proportionality coefficient corresponding to the educational level distribution, and the proportionality coefficient a2 < a1 < a3 corresponding to the educational level distribution; PM represents the maximum personnel density that the management area can bear, Pm represents the personnel density in the management area, Pt represents the number of permanent residents in the management area, and Pl represents the number of floating people in the management area.

[0028] Further, generating the building influence coefficient according to the building life data and living environment data in the building data includes:

[0029] Extracting the building life years ESk and the built years SSk in the building life data, where k is the building number; obtaining the monitoring values HCSnm of each environmental item at each set time period in the living environment data, where n is the environmental item number and m is the set time period number; the environmental items include environmental temperature and environmental humidity, etc.;

[0030] Through the formula calculate the building influence coefficient JY within the management area; where ZCSn is the optimal value corresponding to the environmental project numbered n; DCn is the unit difference corresponding to the environmental project numbered n set; βn is the proportionality coefficient corresponding to the environmental project numbered n, and the specific value is set according to experience; n = 1, 2,..., N; N is the total number of environmental projects; where m = 1, 2,..., M, M represents the total monitoring time; k = 1, 2,..., K; K is the total number of buildings;

[0031] Furthermore, generating the area supervision level according to the personnel influence coefficient and the building influence coefficient includes:

[0032] Calculate the area supervision value R through the formula R = γ1×PA + γ2×JY; generate the area supervision level according to the area supervision value, where γ1 and γ2 represent proportionality coefficients, and the specific values are set according to experience;

[0033] When the area supervision value is within the set low-level monitoring threshold range, set the area supervision level of the corresponding management area to the low-level supervision level;

[0034] When the area supervision value is within the set medium-level monitoring threshold range, set the area supervision level of the corresponding management area to the medium-level supervision level;

[0035] When the area supervision value is within the set high-level monitoring threshold range, set the area supervision level of the corresponding management area to the high-level supervision level;

[0036] The area supervision level includes a low-level supervision level, a medium-level supervision level, and a high-level supervision level.

[0037] Furthermore, generating the security warning signal according to the area supervision level and its corresponding area supervision score, the specific steps include:

[0038] Obtain the area supervision level and area supervision score of the management area; and the upper threshold value and lower threshold value of the supervision score corresponding to the area supervision level;

[0039] Judge whether the area supervision score is greater than the set upper threshold value of the supervision score; if so, generate a supervision surplus signal; if not, then when the area supervision score is less than the set lower threshold value of the supervision score, generate a supervision gap signal; the security warning signal includes a supervision surplus signal and a supervision gap signal.

[0040] The second aspect of this application provides an intelligent monitoring method based on urban security, including the following steps:

[0041] Step 1: Obtain the detection data of each management area;

[0042] Step 2: Extract several regional feature data and monitoring video data from the detection data;

[0043] Step 3: Generate a regional supervision score for the management area according to several monitoring video data;

[0044] Step 4: Generate a corresponding regional supervision level for the management area according to the regional feature data;

[0045] Step 5: Generate a security warning signal according to the regional supervision level and its corresponding regional supervision score; and issue a warning according to the security warning signal.

[0046] Compared with the prior art, the beneficial effects of this application are as follows:

[0047] 1. This application obtains the detection data of each management area through the corresponding connected data acquisition device; extracts several regional feature data and monitoring video data from the detection data; generates the regional supervision score of the management area according to several monitoring video data; generates the corresponding regional supervision level for the management area according to the regional feature data; generates a security warning signal according to the regional supervision level and its corresponding regional supervision score; issues a warning according to the security warning signal; and conducts targeted security monitoring according to the characteristics of each management area, making the distribution of security monitoring in the city more reasonable, thereby improving the accuracy and comprehensiveness of the security intelligent monitoring system in each management area of the city, that is, improving the overall efficiency of urban security.

[0048] 2. This application obtains the regional supervision score corresponding to the management area by averaging the image recognition scores of each monitoring image; the closer the monitoring image corresponding to the image recognition score is to the current moment, the greater its weight value; thereby making it possible to improve the accuracy by adding security devices subsequently.

[0049] 3. This application obtains the detection data of each management area, extracts several regional feature data and monitoring video data from the detection data, generates the regional supervision score of the management area according to several monitoring video data; and generates the corresponding regional supervision level for the management area according to the regional feature data, and then generates a security warning signal according to the regional supervision level and its corresponding regional supervision score; improving the accuracy and comprehensiveness of the security intelligent monitoring system in each management area of the city. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0051] Figure 1 It is a schematic diagram of the system structure of this application;

[0052] Figure 2 It is a schematic diagram of the method steps of this application;

[0053] Figure 3 It is a schematic diagram of the regional supervision level division of this application. Specific implementation manners

[0054] Next, the technical solutions of this application will be clearly and completely described in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this application.

[0055] Please refer to Figure 1 , an embodiment of the first aspect of this application provides a smart monitoring system based on urban security prevention, including: a data collection module, a data processing module, a security warning module, and a database.

[0056] Data collection module: Obtain detection data of each management area through corresponding data collection devices;

[0057] Data processing module: Extract several regional feature data and monitoring video data from the detection data; Generate a regional supervision score for the management area according to several monitoring video data; And generate a corresponding regional supervision level for the management area according to the regional feature data;

[0058] Security warning module: Generate a security warning signal according to the regional supervision level and its corresponding regional supervision score; Give a warning according to the security warning signal.

[0059] In this embodiment, detection data of each management area is obtained through corresponding data collection devices; several regional feature data and monitoring video data are extracted from the detection data; a regional supervision score of the management area is generated according to several monitoring video data; a corresponding regional supervision level for the management area is generated according to the regional feature data; a security warning signal is generated according to the regional supervision level and its corresponding regional supervision score; a warning is given according to the security warning signal; and targeted security monitoring is carried out according to the characteristics of each management area, so that the distribution of security monitoring in the city is more reasonable, thereby improving the accuracy and comprehensiveness of the security smart monitoring system in each management area of the city, that is, improving the overall efficiency of urban security prevention.

[0060] The data acquisition device includes: monitoring camera devices, temperature sensors, etc.; the monitoring device includes cameras, etc.; the management area can be divided according to the functional division method or the street division method, etc.; according to the functional division method, it includes commercial areas, residential areas, and industrial areas; the street division method specifically divides the management area into different streets, and the specific streets are set according to experience;

[0061] The detected data includes the regional feature data and monitoring video data of the management area; the monitoring video data is frame-extracted according to the set frame extraction rule to obtain a number of monitoring images; the regional feature data includes personnel data and building data; the regional supervision score is to evaluate the quality of the security monitoring effect of the management area;

[0062] Specifically, generating the regional supervision score of the management area according to a number of monitoring video data includes:

[0063] Obtain a number of monitoring video data, and frame-extract the monitoring video data according to the set frame extraction rule to obtain a number of monitoring images; input the monitoring images into the image evaluation model to obtain the image recognition score corresponding to the monitoring images; generate the regional supervision score according to the image recognition scores corresponding to each monitoring image.

[0064] In this embodiment, specifically, it should be noted that the frame extraction rule is to extract each image frame in the monitoring video data, which is obtained by converting the video data into image frames according to the content recorded in line

[0031] of the patent specification of the invention named "A Smart City Security Monitoring System Based on Image Recognition Technology" with the publication number CN117522017B.

[0065] The image evaluation model is obtained by training an artificial intelligence model and includes:

[0066] Obtain a number of monitoring images and their corresponding image recognition scores through the database; the image recognition score is the evaluation made by experts on the monitoring images according to the clarity and color presentation ability of the monitoring images, etc. The higher the clarity of the monitoring image and the better the color presentation ability, it indicates that the quality of the video data collected by the corresponding monitoring device is higher, the result of its recognition is more accurate, and the corresponding image recognition score is set larger; integrate the monitoring images and their corresponding image recognition scores into a number of training data and test data;

[0067] The color presentation ability is mainly measured by factors such as color gamut, contrast, brightness, white balance, and color temperature; the values are set according to experience.

[0068] Train an artificial intelligence model using training data, and test the trained artificial intelligence model using test data; specifically, input the monitoring images and their corresponding image recognition scores in the test data into the trained artificial intelligence model, and output the image recognition scores. Determine whether the difference between the image recognition scores and the image recognition scores recorded in the test data is within an acceptable range; if yes, it means that this set of test data passes the test, and continue to test the next set of test data; if not, the relevant parameters of the artificial intelligence model need to be adjusted, and continue to use this set of test data for testing; until a set percentage of the test data passes the test; finally, obtain an image evaluation model with the monitoring images and their corresponding image recognition scores as input and the image recognition scores as output; where the artificial intelligence model is a convolutional neural network model.

[0069] Specifically, generate a regional supervision score according to the image recognition scores corresponding to each monitoring image, including:

[0070] Obtain the image recognition scores of each monitoring image and label them as TX ij ; where i is the number of the monitoring device corresponding to the monitoring image, and j is the frame number corresponding to the monitoring image;

[0071] Through the formula Calculate the regional supervision score TA corresponding to the management area; where j = 1, 2,..., J, J represents the total number of frames corresponding to the monitoring image; α i Represents the weight coefficient of the corresponding monitoring image; i = 1, 2,..., F, F represents the total number of monitoring devices corresponding to the management area; in this embodiment, the regional supervision scores corresponding to each management area are calculated through the above formula. When there are more monitoring devices in the corresponding management area and the current status of each monitoring device is better, that is, when the several image recognition scores corresponding to the video data collected by each monitoring device recently are higher, it means that the security monitoring effect of the corresponding area is better, and the corresponding regional supervision score is set larger; on the contrary, when there are fewer monitoring devices in the corresponding management area and the current status of each monitoring device is worse, that is, when the several image recognition scores corresponding to the video data collected by each monitoring device recently are lower, it means that the security monitoring effect of the corresponding area is worse, and the corresponding regional supervision score is set smaller.

[0072] Among them, Among them, f(i) represents an increasing function; the area supervision score corresponding to the management area is obtained by averaging the image recognition scores of each monitoring image through a formula; the closer the monitoring image corresponding to the image recognition score is to the size of the monitoring area, the greater its weight value; in addition, in this embodiment, f(i) = log(1 + i + c), where c represents a constant and is greater than 0; f(i) can also be specifically set to other increasing functions of i according to experience, and the weight coefficient α corresponding to different increasing functions i is different.

[0073] In this application, the area supervision score corresponding to the management area is obtained by averaging the image recognition scores of each monitoring image; the closer the monitoring image corresponding to the image recognition score is to the current moment, the greater its weight value; thereby improving the accuracy by adding security equipment subsequently.

[0074] Specifically, generating the area supervision level corresponding to the management area according to the area feature data includes:

[0075] Extracting the personnel data and building data in the area feature data;

[0076] Generating a personnel influence coefficient according to the number of permanent residents, the number of floating population, the population density, and the educational level in the personnel data;

[0077] Generating a building influence coefficient according to the building life data and the living environment data in the building data;

[0078] Generating the area supervision level according to the personnel influence coefficient and the building influence coefficient.

[0079] Specifically, the generating the personnel influence coefficient according to the number of permanent residents, the number of floating population, the population density, and the educational level distribution includes:

[0080] Extracting the number of permanent residents, the number of floating population, and the population density of the management area in the personnel data; the educational level distribution of the management area specifically includes: D1 is the number of people receiving compulsory education in the management area, D2 is the number of people receiving higher education in the management area, and D3 is the number of people not receiving education in the management area; D z is the number of people corresponding to the educational level distribution numbered z; z = 1, 2, 3;

[0081] Through the formula Calculate to obtain the personnel influence coefficient PA within the management area; a zIt represents the proportionality coefficient corresponding to the educational attainment distribution, and the proportionality coefficients a2 < a1 < a3 corresponding to the educational attainment distribution; the specific values are set according to experience; PM represents the maximum population density that the management area can bear, Pm represents the population density of the management area, Pt represents the number of permanent residents in the management area, Pl represents the number of floating people in the management area, and Pt > Pl.

[0082] The number of permanent residents includes people who live there all year round and people with local household registrations, etc.; the number of floating people includes tourists, business travelers, etc.

[0083] In this embodiment, the relationship between the personnel influence coefficient of each management area and the population density of the management area, the number of non-local people in the management area, the number of local people in the management area, and the educational attainment distribution is obtained through formula calculation; when the population density of the management area is greater and the number of non-local people is greater, it indicates that the number of people in the management area is larger and the personnel composition is more complex, and the probability of dangerous accidents is higher, and the corresponding personnel influence coefficient is larger; when the number of people with a high educational attainment in the management area is larger, it indicates that the quality of the personnel in the management area is higher, and they have stronger comprehension and communication abilities, the probability of dangerous situations is lower, and the ability of on-site personnel to assist in handling dangerous situations is also stronger, and the corresponding personnel influence coefficient is smaller; vice versa.

[0084] Specifically, a building influence coefficient is generated based on the building life data and living environment data in the building data, including:

[0085] Extract the building life years ESk and the years since the building was completed SSk in the building life data, where k is the building number; obtain the monitoring values HCSnm of each environmental item in each setting time period in the living environment data, where n is the environmental item number and m is the setting time period number; the environmental items include environmental temperature, environmental humidity, etc.

[0086] Through the formula Calculate the building influence coefficient JY within the management area; where ZCSn is the optimal value corresponding to the environmental item numbered n; DCn is the unit difference corresponding to the environmental item numbered n set; βn is the proportionality coefficient corresponding to the environmental item numbered n, and the specific value is set according to experience; n = 1, 2,..., N; N is the total number of environmental items; where m = 1, 2,..., M, M represents the total monitoring time; k = 1, 2,..., K; K is the total number of buildings.

[0087] In this embodiment, it should be specifically noted that the buildings specifically include civilian buildings, commercial buildings, office buildings, etc.; the relationship between the building influence coefficients of each management area and the monitoring values of each environmental item in each setting time period and the built years of the buildings is obtained through formula calculation; when the monitoring values of each environmental item in each setting time period deviate more from the optimal values, it indicates that the buildings in the management area are more affected by environmental factors, the aging degree of the corresponding buildings is more serious, and the corresponding building influence coefficient is larger; when the built years of the buildings are larger, it indicates that the aging degree of the buildings in the management area is higher, and the corresponding building influence coefficient is larger; on the contrary, the corresponding building influence coefficient is smaller.

[0088] Specifically, the area supervision level is generated according to the personnel influence coefficient and the building influence coefficient, including:

[0089] The area supervision value R is calculated through the formula R = γ1×PA + γ2×JY; the area supervision level is generated according to the area supervision value, where γ1 and γ2 represent proportionality coefficients used to adjust the proportional relationship between the two parameters; the specific values are set according to experience;

[0090] When the area supervision value is within the set low-level monitoring threshold range, the area supervision level of the corresponding management area is set to the low-level supervision level;

[0091] When the area supervision value is within the set medium-level monitoring threshold range, the area supervision level of the corresponding management area is set to the medium-level supervision level;

[0092] When the area supervision value is within the set high-level monitoring threshold range, the area supervision level of the corresponding management area is set to the high-level supervision level; it can be understood that the management areas corresponding to the high-level supervision level require more security equipment or patrol personnel than those corresponding to the medium-level supervision level; the management areas corresponding to the medium-level supervision level require more security equipment or patrol personnel than those corresponding to the low-level supervision level; in this embodiment, only an example is given that the area supervision is divided into three levels, and in other embodiments, it can be divided into several levels. The more detailed the level division is, the stronger the pertinence of supervising different management areas will be, and it can better improve the accuracy and comprehensiveness of the security intelligent monitoring system for each management area in the city, and improve the overall efficiency of urban security.

[0093] Please refer to Figure 3 , and the area supervision level includes the low-level supervision level, the medium-level supervision level, and the high-level supervision level.

[0094] Specifically, the security warning signal is generated according to the area supervision level and its corresponding area supervision score. The specific steps include:

[0095] A1: Obtain the area supervision level and area supervision score of the management area; as well as the upper threshold value and lower threshold value of the supervision score corresponding to the area supervision level;

[0096] A2: Determine whether the area supervision score is greater than the set upper threshold value of the supervision score; if so, generate a supervision surplus signal; if not, when the area supervision score is less than the set lower threshold value of the supervision score, generate a supervision gap signal; the security warning signal includes a supervision surplus signal and a supervision gap signal; it can be understood that the upper threshold value and lower threshold value of the supervision score corresponding to different supervision levels are different, the upper threshold value of the supervision score of the high supervision level is greater than the upper threshold value of the supervision score of the medium supervision level; the upper threshold value of the supervision score of the medium supervision level is greater than the upper threshold value of the supervision score of the low supervision level; the lower threshold value of the supervision score of the high supervision level is greater than the lower threshold value of the supervision score of the medium supervision level; the lower threshold value of the supervision score of the medium supervision level is greater than the lower threshold value of the supervision score of the low supervision level.

[0097] In this embodiment, the operations of steps A1 - A2 are performed according to the area supervision level corresponding to each management area; when the management area generates a supervision surplus signal, it indicates that there are too many security devices in the management area, which means that the security devices or patrol personnel in this management area should be reduced accordingly; when the management area generates a supervision gap signal, it indicates that there is a shortage of security devices in the management area, and the security devices or patrol personnel in the management area should be increased accordingly.

[0098] Please refer to Figure 2 , the second aspect of this application provides an intelligent monitoring method based on urban security, including the following steps:

[0099] Step 1: Obtain the detection data of each management area;

[0100] Step 2: Extract several area feature data and monitoring video data from the detection data;

[0101] Step 3: Generate the area supervision score of the management area according to several monitoring video data;

[0102] Step 4: Generate the area supervision level corresponding to the management area according to the area feature data;

[0103] Step 5: Generate a security warning signal according to the area supervision level and its corresponding area supervision score; and give a warning according to the security warning signal.

[0104] The working principle of this application:

[0105] This application obtains the detection data of each management area through a data acquisition device connected correspondingly; extracts several regional feature data and monitoring video data from the detection data; generates a regional supervision score for the management area according to the several monitoring video data; generates a corresponding regional supervision level for the management area according to the regional feature data; generates a security warning signal according to the regional supervision level and its corresponding regional supervision score; issues a warning according to the security warning signal; and conducts targeted security monitoring according to the characteristics of each management area, making the distribution of security monitoring in the city more reasonable, thereby improving the accuracy and comprehensiveness of the security intelligent monitoring system in each management area of the city, that is, improving the overall efficiency of urban security.

[0106] The above embodiments are only used to illustrate the technical method of this application and not to limit it. Although this application has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of this application can be modified or equivalently replaced without departing from the spirit and scope of the technical method of this application.

Claims

1. An intelligent monitoring system based on urban security, characterized in that: Data collection module: obtains detection data of each management area; Data processing module: extracting several regional feature data and surveillance video data from the detection data; generating a regional supervision score for the management area based on several surveillance video data; and generating a regional supervision level corresponding to the management area based on the regional feature data; Security warning module: Generate security warning signals according to the regional supervision level and its corresponding regional supervision score; issue warnings based on the security warning signals.

2. According to claim 1, a smart monitoring system based on urban security is characterized in that: Generate a regional supervision score for a management area based on a number of surveillance video data, including: Acquire a number of monitoring video data, extract frames from the monitoring video data according to the set frame extraction rules to obtain a number of monitoring images; input the monitoring images into the image evaluation model to obtain image recognition scores corresponding to the monitoring images; and generate regional supervision scores according to the image recognition scores corresponding to the respective monitoring images.

3. According to claim 2, a smart monitoring system based on urban security is characterized in that: The image evaluation model is obtained through artificial intelligence model training, including: The database obtains a number of monitoring images and their corresponding image recognition scores; the monitoring images and their corresponding image recognition scores are integrated into a number of training data and test data; The artificial intelligence model is trained using training data, and the trained artificial intelligence model is tested using test data, ultimately obtaining an image evaluation model whose input is a monitoring image and its corresponding image recognition score, and whose output is an image recognition score, wherein the artificial intelligence model is a convolutional neural network model.

4. According to claim 2, the intelligent monitoring system based on urban security is characterized in that: The generating of the regional supervision score according to the image recognition score corresponding to each monitoring image includes: Get the image recognition score of each monitoring image and mark it as TX ij ; Wherein, i is the number of the monitoring device corresponding to the monitoring image, and j is the frame number corresponding to the monitoring image; By formula Calculate and obtain the regional supervision score TA corresponding to the management area; wherein j=1, 2, ..., J, J represents the total number of frames corresponding to the monitoring image; α i Represents the weight coefficient of the corresponding monitoring image; i=1, 2, ..., F, F represents the total number of monitoring devices corresponding to the management area.

5. The intelligent monitoring system based on urban security according to claim 1 is characterized in that: The generating of the regional supervision level corresponding to the management area according to the regional characteristic data includes: Extracting personnel data and building data from regional feature data; Generate the personnel impact coefficient based on the number of local personnel, number of foreign personnel, personnel density and educational level in the personnel data; Generate building impact coefficients based on building life data and living environment data in building data; Generate regional supervision level based on personnel impact coefficient and building impact coefficient.

6. The intelligent monitoring system based on urban security according to claim 5 is characterized in that: The personnel impact coefficient is generated according to the number of permanent residents, number of floating personnel, personnel density and educational level in the personnel data, including: According to the number of permanent residents, floating population, and population density in the management area in the personnel data; the distribution of educational level in the management area specifically includes: D1 is the number of people receiving compulsory education in the management area, D2 is the number of people receiving higher education in the management area, and D3 is the number of people not receiving education in the management area; D z is the number of people corresponding to the educational level distribution numbered z; z = 1, 2, 3; Through the formula The personnel influence coefficient PA within the management area is calculated; a z represents the proportionality coefficient corresponding to the educational level distribution, and the proportionality coefficient a2 < a1 < a3 corresponding to the educational level distribution; PM represents the maximum personnel density that the management area can bear, Pm represents the personnel density of the management area, Pt represents the number of local personnel in the management area, and Pl represents the number of non-local personnel in the management area.

7. The intelligent monitoring system based on urban security according to claim 5 is characterized in that: The building impact coefficient is generated according to the building life data and the living environment data in the building data, including: Extract the building lifespan ESk and the building construction years SSk from the building lifespan data, where k is the building number; obtain the monitoring values ​​HCSnm of each environmental item in the residential environment data in each set time period, where n is the number of the environmental item and m is the number of the set time period; By formula The building impact coefficient JY within the management area is calculated; wherein, ZCSn is the optimal value corresponding to the environmental item numbered n; DCn is the unit difference corresponding to the set environmental item numbered n; βn is the proportional coefficient corresponding to the environmental item numbered n; n=1, 2, ..., N; N is the total number of environmental items; wherein m=1, 2, ..., M, M represents the total monitoring time; k=1, 2, ..., K; K is the total number of buildings.

8. The intelligent monitoring system based on urban security according to claim 5 is characterized in that: The regional supervision level is generated according to the personnel impact coefficient and the building impact coefficient, including: The regional supervision value R is calculated by the formula R=γ1×PA+γ2×JY; the regional supervision level is generated according to the regional supervision value, and γ1 and γ2 represent proportional coefficients; the regional supervision level includes a low supervision level, a medium supervision level and a high supervision level; When the regional supervision value is within the set low-level monitoring threshold range, the regional supervision level of the corresponding management area is set to the low-level supervision level; When the regional supervision value is within the set intermediate monitoring threshold range, the regional supervision level of the corresponding management area is set to the intermediate supervision level; When the regional supervision value is within the set advanced monitoring threshold range, the regional supervision level of the corresponding management area is set to the advanced supervision level.

9. The intelligent monitoring system based on urban security according to claim 1 is characterized in that: The security warning signal is generated according to the regional supervision level and its corresponding regional supervision score, and the specific steps include: Obtain the regional supervision level and regional supervision score of the management area; as well as the upper and lower supervision score thresholds corresponding to the regional supervision level; Determine whether the regional supervision score is greater than the set supervision score upper limit threshold; if yes, generate a supervision excess signal; if no, when the regional supervision score is less than the set supervision score lower limit threshold, generate a supervision gap signal; the security warning signal includes a supervision excess signal and a supervision gap signal.

10. A smart monitoring method based on urban security, applied to the operation of a smart monitoring system based on urban security as described in any one of claims 1 to 9; characterized in that: The following steps are involved: Step 1: Obtain the detection data of each management area; Step 2: Extracting some regional feature data and surveillance video data from the detection data; Step 3: Generate a regional supervision score for the management area based on a number of surveillance video data; Step 4: Generate the regional supervision level corresponding to the management area based on the regional characteristic data; Step 5: Generate a security warning signal based on the regional supervision level and its corresponding regional supervision score; and issue a warning based on the security warning signal.

Images