WO2021027569A1 - 执法分析方法、装置、电子设备及存储介质 - Google Patents

执法分析方法、装置、电子设备及存储介质 Download PDF

Info

Publication number
WO2021027569A1
WO2021027569A1 PCT/CN2020/105341 CN2020105341W WO2021027569A1 WO 2021027569 A1 WO2021027569 A1 WO 2021027569A1 CN 2020105341 W CN2020105341 W CN 2020105341W WO 2021027569 A1 WO2021027569 A1 WO 2021027569A1
Authority
WO
WIPO (PCT)
Prior art keywords
reconsideration
illegal
law enforcement
change index
index
Prior art date
Application number
PCT/CN2020/105341
Other languages
English (en)
French (fr)
Inventor
党升
曹思佳
Original Assignee
平安国际智慧城市科技股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 平安国际智慧城市科技股份有限公司 filed Critical 平安国际智慧城市科技股份有限公司
Publication of WO2021027569A1 publication Critical patent/WO2021027569A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/12—Computing arrangements based on biological models using genetic models
    • G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/18—Legal services

Definitions

  • This application relates to the field of data mining technology, and more specifically, to a law enforcement analysis method, device, electronic equipment, and storage medium.
  • Civilized law enforcement is to establish the concept of people-oriented, law-based administration, and governance for the people in administrative law enforcement, fully respect the rights and interests of administrative law enforcement counterparts, strictly follow law enforcement procedures, adhere to the combination of education and punishment, and combine management and service. Continuously improve the efficiency of administrative law enforcement to provide guarantee for the construction of a harmonious society and a legal society.
  • the purpose of this application is to provide a law enforcement analysis method, device, electronic equipment and storage medium that objectively analyze whether law enforcement is civilized or healthy.
  • this application provides a law enforcement analysis method, including:
  • Step S1 Send the event pictures collected by the law enforcement terminal to the server, extract image features from the event pictures and determine whether they are illegal, and collect the number of events judged to be illegal at different moments to form a data set of illegal amounts.
  • the server collects the number of events that are determined to be illegal and proposes reconsideration, and obtains a reconsideration amount data set composed of reconsideration amounts at different moments, the law enforcement terminal includes a law enforcement recorder and a mobile terminal with a shooting function;
  • Step S2 Obtain the reconsideration rate data set through the illegal amount data set and the reconsideration amount data set uploaded to the server according to the following formula:
  • F t is the illegal amount at time t
  • M t is the amount of reconsideration at time t
  • p t is the reconsideration rate at time t
  • Step S3 The server obtains the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index at each moment through the illegal amount data set, the reconsideration amount data set, and the reconsideration rate data set;
  • Step S4 the server obtains law enforcement health through the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index, where the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index are negatively correlated with law enforcement health .
  • this application also provides a law enforcement analysis device, including a law enforcement terminal and a server.
  • the law enforcement terminal includes a law enforcement recorder and a mobile terminal with a shooting function.
  • the server includes an illegal amount acquisition unit and a reconsideration amount. Acquisition unit, reconsideration rate acquisition unit, law enforcement change index acquisition unit and law enforcement analysis unit, including:
  • the law enforcement terminal collects event pictures and sends them to the server;
  • the illegal quantity obtaining unit extracts image features and judges whether it is illegal, the sum of illegal events at one moment is regarded as the illegal quantity, and the illegal quantity at different times constitutes the illegal quantity data set;
  • the reconsideration amount obtaining unit uses the sum of the events that are judged to be illegal at one time and proposes reconsideration as the reconsideration amount, and the reconsideration amount data set constituted by the reconsideration amounts at different moments;
  • the reconsideration rate obtaining unit obtains the illegal amount data set obtained by the illegal amount obtaining unit and the reconsideration amount data set obtained by the reconsideration amount obtaining unit according to the following formula to obtain the reconsideration rate data set
  • F t is the illegal amount at time t
  • M t is the amount of reconsideration at time t
  • p t is the reconsideration rate at time t
  • the law enforcement change index obtaining unit obtains the illegal incremental change index, the reconsideration incremental change index, and the reconsideration rate change index at each moment through the illegal amount data set, the reconsideration amount data set, and the reconsideration rate data set;
  • the law enforcement analysis unit obtains law enforcement health through the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index.
  • the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index are negative for law enforcement health. Related.
  • the present application also provides an electronic device, including a memory and a processor, the memory stores a law enforcement analysis program, and when the law enforcement analysis program is executed by the processor, the foregoing law enforcement analysis method is implemented A step of.
  • the present application also provides a computer-readable storage medium that includes a law enforcement analysis program, which when executed by a processor, implements the above-mentioned law enforcement analysis method step.
  • the law enforcement analysis method, device, electronic equipment, and computer-readable storage medium described in this application can intuitively understand the health of law enforcement through quantified values.
  • the illegal incremental change index, the reconsideration incremental change index, and the reconsideration rate change index are all Incremental information is added, not only the law enforcement quantity but also the law enforcement index change direction, positive or negative information, to objectively analyze the health of law enforcement, and through the influencing factors of each indicator, we can know the specific factors affecting the decline in health for correlation Departmental decision.
  • FIG. 1 is a schematic diagram of the application environment of a preferred embodiment of the law enforcement analysis method of this application;
  • FIG. 2 is a schematic diagram of modules of a preferred embodiment of the law enforcement analysis program in FIG. 1;
  • Fig. 3 is a flowchart of a preferred embodiment of the law enforcement analysis method of the present application.
  • This application provides a law enforcement analysis method, which is applied to an electronic device 1.
  • FIG. 1 it is a schematic diagram of the application environment of the preferred embodiment of the law enforcement analysis method of this application.
  • the electronic device 1 may be a terminal client with computing functions, such as a server, a mobile phone, a tablet computer, a portable computer, a desktop computer, and the like.
  • the memory 11 includes at least one type of readable storage medium.
  • the at least one type of readable storage medium may be a non-volatile storage medium such as flash memory, hard disk, multimedia card, card-type memory, and the like.
  • the readable storage medium may be an internal storage unit of the electronic device 1, such as a hard disk of the electronic device 1.
  • the readable storage medium may also be an external memory of the electronic device 1, for example, a plug-in hard disk or a smart memory card (Smart Media Card, SMC) equipped on the electronic device 1. Secure Digital (SD) card, flash card (Flash Card), etc.
  • SD Secure Digital
  • flash card Flash Card
  • the readable storage medium of the memory 11 is generally used to store the law enforcement analysis program 10 and the like installed in the electronic device 1.
  • the memory 11 can also be used to temporarily store data that has been output or will be output.
  • the processor 12 may be a central processing unit (CPU), microprocessor or other data processing chip in some embodiments, and is used to run program codes or process data stored in the memory 11, for example, execute a law enforcement analysis program 10 and so on.
  • CPU central processing unit
  • microprocessor or other data processing chip in some embodiments, and is used to run program codes or process data stored in the memory 11, for example, execute a law enforcement analysis program 10 and so on.
  • the network interface 13 can optionally include a standard wired interface and a wireless interface (such as a Wi-Fi interface), and is usually used to establish a communication connection between the electronic device 1 and other electronic clients.
  • a standard wired interface and a wireless interface such as a Wi-Fi interface
  • the communication bus 14 is used to realize the connection and communication between these components.
  • FIG. 1 only shows the electronic device 1 with the components 11-14, but it should be understood that it is not required to implement all the illustrated components, and more or fewer components may be implemented instead.
  • the electronic device 1 may also include a user interface.
  • the user interface may include an input unit such as a keyboard (Keyboard), a voice input device such as a microphone (microphone) and other clients with voice recognition functions, and a voice output device such as a speaker or earphone.
  • the user interface may also include a standard wired interface and a wireless interface.
  • the electronic device 1 may also include a display, and the display may also be called a display screen or a display unit.
  • it may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an organic light-emitting diode (Organic Light-Emitting Diode, OLED) touch device.
  • the display is used for displaying information processed in the electronic device 1 and for displaying a visualized user interface.
  • the electronic device 1 further includes a touch sensor.
  • the area provided by the touch sensor for the user to perform a touch operation is called a touch area.
  • the touch sensor described here may be a resistive touch sensor, a capacitive touch sensor, or the like.
  • the touch sensor includes not only a contact type touch sensor, but also a proximity type touch sensor and the like.
  • the touch sensor may be a single sensor, or may be, for example, a plurality of sensors arranged in an array.
  • the electronic device 1 may also include logic gate circuits, sensors, audio circuits, etc., which will not be repeated here.
  • the memory 11 as a computer storage medium may include an operating system and a law enforcement analysis program 10; when the processor 12 executes the law enforcement analysis program 10 stored in the memory 11, the following steps are implemented:
  • Step S1 Send the event pictures collected by the law enforcement terminal to the server, extract image features from the event pictures and determine whether they are illegal, and collect the number of events judged to be illegal at different moments to form a data set of illegal amounts.
  • the server collects the number of events that are determined to be illegal and proposes reconsideration, and obtains a reconsideration amount data set composed of reconsideration amounts at different moments.
  • the law enforcement terminal includes law enforcement recorders and mobile terminals (mobile phones, cameras, etc.) with shooting functions, such as In actual law enforcement, law enforcement officers upload illegal incidents to the server through the law enforcement recorder or mobile phone through the network to constitute a data set of illegal quantities.
  • the illegal data and records will be notified to the illegal personnel through email, text message, telephone, etc., and the illegal personnel have disputes in the violation of the law.
  • the reconsideration can be filed online through the official website of law enforcement records, or through the manual window, the reconsideration data will be uploaded to the server to form a reconsideration amount data set;
  • Step S2 Obtain the reconsideration rate data set according to the following formula (1) through the illegal amount data set and the reconsideration amount data set uploaded to the server
  • F t is the illegal amount at time t
  • M t is the amount of reconsideration at time t
  • p t is the reconsideration rate at time t
  • Step S3 the server obtains the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index at each time according to the following formulas (2)-(4) through the illegal amount data set, reconsideration amount data set, and reconsideration amount data set.
  • N is the time period set before the current moment
  • Z M is the average illegal amount in the time period
  • Z f is the average reconsideration amount in the time period
  • Z p is the average reconsideration rate in the time period
  • max( M N ) is the maximum illegal amount in the time period
  • min(M N ) is the minimum illegal amount in the time period
  • max(f N ) is the maximum reconsideration amount in the time period
  • min(f N ) is the The minimum amount of reconsideration in the time period
  • max(p N ) is the maximum reconsideration rate in the time period
  • min(p N ) is the minimum reconsideration rate in the time period
  • B M is the illegal incremental change index
  • B f is Reconsideration increment change index
  • B p is the reconsideration rate change index
  • Step S4 the server obtains law enforcement health through the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index, where the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index are negatively correlated with law enforcement health That is to say, the greater the illegal incremental change index, the reconsideration incremental change index, and the reconsideration rate change index, the lower the law enforcement health, where the law enforcement health is a measure of the health of the administrative subject’s law enforcement (civilized law enforcement) Degree).
  • step S3 when the processor 12 executes the law enforcement analysis program 10 stored in the memory 11, the following steps are further implemented: between step S3 and step S4, it further includes: a change index for illegal increments, a reconsideration increment change index, and a reconsideration rate change index Perform a weighted combination according to the following formula (5) to obtain a comprehensive index
  • Z is the composite index
  • the comprehensive index is positively correlated with the health of law enforcement, that is, as the comprehensive index is larger, the health of law enforcement is higher, for example, the comprehensive index has a value range [-1, 1], and the value Sign conversion, Z greater than 0 indicates good law enforcement health, Z less than 0 indicates low law enforcement health, Z and 0 indicate normal law enforcement health.
  • the law enforcement analysis program 10 may also be divided into one or more modules, and the one or more modules are stored in the memory 11 and executed by the processor 12 to complete the application.
  • the module referred to in this application refers to a series of computer program instruction segments that can complete specific functions.
  • FIG. 2 it is a functional block diagram of a preferred embodiment of the law enforcement analysis program 10 in FIG. 1.
  • the law enforcement analysis program 10 can be divided into a collection module 110 and a server 120.
  • the server 120 includes an illegal amount obtaining unit 121, a reconsideration amount obtaining unit 122, a reconsideration rate obtaining unit 123, a law enforcement change index obtaining unit 124, and a law enforcement analysis unit 125.
  • the collection module 110 sends the event pictures collected by the law enforcement terminal to the server 120;
  • the illegal quantity obtaining unit 121 extracts image features and judges whether it is illegal, the sum of the illegal events at a time is regarded as the illegal quantity, and the illegal quantity at different times constitutes Illegal amount data set;
  • the reconsideration amount obtaining unit 122 takes the sum of the events that are determined to be illegal at one moment and submits reconsideration as the amount of reconsideration, and the amount of reconsideration data sets constituted by the amount of reconsideration at different moments;
  • the reconsideration rate obtaining unit 123 passes the illegal
  • the illegal amount data set obtained by the amount obtaining unit 121 and the reconsideration amount data set obtained by the reconsideration amount obtaining unit 122 obtain the reconsideration rate data set;
  • the law enforcement change index obtaining unit 124 obtains the reconsideration rate data set through the illegal amount data set, the reconsideration amount data set, and the reconsideration rate data
  • the law enforcement analysis unit 126 obtains law enforcement health through the illegal incremental change index, reconsideration incremental change index
  • the server 120 further includes a comprehensive index obtaining unit 125 to perform a weighted combination of the illegal incremental change index, the reconsideration incremental change index, and the reconsideration rate change index according to the following formula to obtain the comprehensive index.
  • the law enforcement analysis unit analyzes according to the comprehensive index The health of law enforcement, as the comprehensive index is larger, the health of law enforcement is higher.
  • this application also provides a law enforcement analysis method.
  • Figure 3 is a flowchart of a preferred embodiment of the law enforcement analysis method of this application.
  • the method can be executed by a device, and the device can be implemented by software and/or hardware.
  • the law enforcement analysis method includes:
  • Step S1 the event pictures collected by the law enforcement terminal are sent to the server to extract the image features of the event pictures and judge whether they are illegal, the number of incidents judged as illegal at different moments is collected to form a data set of illegal quantity, and the server collects The number of incidents that are determined to be illegal and the reconsideration is raised, and the reconsideration amount data set composed of the amount of reconsideration at different times is obtained.
  • the law enforcement terminal includes a law enforcement recorder and a mobile terminal (mobile phone, camera, etc.) with a shooting function;
  • Step S2 Obtain the reconsideration rate data set according to the following formula (1) through the illegal amount data set and the reconsideration amount data set uploaded to the server
  • F t is the illegal amount at time t
  • M t is the amount of reconsideration at time t
  • p t is the reconsideration rate at time t
  • Step S3 the server obtains the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index at each time according to the following formulas (2)-(4) through the illegal amount data set, reconsideration amount data set, and reconsideration amount data set.
  • N is the time period set before the current moment
  • Z M is the average illegal amount in the time period
  • Z f is the average reconsideration amount in the time period
  • Z p is the average reconsideration rate in the time period
  • max( M N ) is the maximum illegal amount in the time period
  • min(M N ) is the minimum illegal amount in the time period
  • max(f N ) is the maximum reconsideration amount in the time period
  • min(f N ) is the The minimum reconsideration amount in the time period
  • max(p N ) is the maximum reconsideration rate in the time period
  • min(p N ) is the minimum reconsideration rate in the time period
  • Step S4 the server obtains law enforcement health through the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index, where the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index are negatively correlated with law enforcement health
  • the larger the illegal incremental change index, the reconsideration incremental change index, and the reconsideration rate change index the lower the law enforcement health.
  • step S1 the method for obtaining the illegal amount includes:
  • the image analysis server is composed of high-end artificial intelligence image recognition technology, which extracts image features through image recognition algorithms;
  • the above image features are input into a convolutional neural network to obtain the illegal probability of the event corresponding to the picture.
  • a convolutional neural network such as VGGNet, AlexNet, or GoogleNet is used to train a classification model of illegal pictures using a convolutional deep learning method.
  • the training of the classification model needs to mark a batch of illegal photos in advance, mix with other photos, and train the classifier through the model to classify whether the photos are illegal photos.
  • the classification model predicts the input photo and the output photo’s illegal probability or illegal status.
  • the convolutional neural network input image size 224x224 pixel RGB image after multi-layer 3D convolution and maximum pooling operation, and then add the neural network fully connected layer , And finally by mapping the picture to a 1000-dimensional feature vector, the feature vector is used as the input of the two classifiers (logistic regression, decision tree, random forest, support vector machine) to train the classifier, the classifier can be fully connected to the two-dimensional by the neural network
  • the softmax function as the objective function for mapping, and output the final result of the illegal probability value, which is a decimal of 0-1.
  • the larger the value the higher the probability of illegality.
  • the illegal facts are confirmed.
  • Machine learning algorithms can also be used. Models such as logist, svm, and Tree perform two classifications and output the probability of violation;
  • the image and the corresponding illegal information and illegal status are recorded in the database of the server to form a data set of illegal amount.
  • the illegal information includes Information on illegal locations, illegal personnel, law enforcement officers, people who report to the public, contact information, illegal photos, illegal status, etc.
  • step S1 the method for obtaining the reconsideration amount data set and the illegal amount data set includes:
  • the illegal data set can be sent to the offender via SMS, email, WeChat, or official account, and the offender will pass when there is a dispute over the illegal information.
  • Official account manual acceptance window, government official website and other channels for reconsideration;
  • a weighted sum of the probability of violation of the violations, the number of violations recorded by the violation, the severity of violations, and the credit rating of the violations for which the reconsideration has not been obtained, will be obtained to obtain the violation confidence of the violations;
  • step S3 and step S4 it further includes: weighted combination of the illegal incremental change index, the reconsideration incremental change index, and the reconsideration rate change index according to the following formula (5) to obtain a comprehensive index
  • Z is the comprehensive index
  • step S4 the comprehensive index is positively correlated with the law enforcement health, that is, as the comprehensive index is larger, the law enforcement health is higher.
  • step S4 includes:
  • mapping function to map the comprehensive index to obtain the law enforcement health according to the following formula (6)
  • x [-1,1]
  • E law enforcement health
  • law enforcement health is positively correlated with law enforcement health.
  • step S3 the method further includes:
  • the abnormal value detection of illegal incremental change index, reconsideration incremental change index, reconsideration rate change index or/and comprehensive index through normal distribution including:
  • X i is the value at the i-th moment of one of the illegal incremental change index, reconsideration incremental change index, reconsideration rate change index, and composite index
  • ⁇ is the average value of the index
  • ⁇ is the standard deviation of the index
  • the illegal incremental change index, the reconsideration incremental change index, the reconsideration rate change index, and the comprehensive index are the result of one index after being discriminated, which is a true value (normal value) or a false value (abnormal value). Valued means true and false results.
  • the above-mentioned outlier detection eliminates particularly large non-conforming values to prevent the overall calculation results from being too small due to certain outliers, which are concentrated in a small data interval, which is not convenient for subsequent calculations, and it is difficult to reflect the results The difference between.
  • step S3 and step S4 it further includes the step of optimizing the weights of the illegal incremental change index, the reconsideration incremental change index, and the reconsideration rate change index using a genetic algorithm.
  • the steps include:
  • the individual genes of each individual are used as the weights of the illegal incremental change index, the reconsideration incremental change index, and the reconsideration rate change index. According to the comprehensive index of the sample belonging to each topic category, the fitness of each individual is obtained, where:
  • a selection strategy based on the fitness ratio is used to select individuals in the initial population, and the selected individuals G u are obtained ;
  • a single-point crossover operator is used to perform cross update on selected individuals.
  • the maximum value of each gene after update is used as the upper bound of the gene, and the minimum value of each gene after update is used as the gene The lower bound
  • the mutation operation is performed on the selected individuals that have undergone cross-update, and the mutated individuals are obtained, which are substituted into the individual evaluation subunit to evolve the initial population, where:
  • g j is the j-th gene of the individual G u selected
  • g jmax and g jmin are the upper and lower bounds of the gene g j
  • r p is the pseudo-random number generated P-th when the individual G u is selected, iter now is the current evolutionary algebra, iter max is the set maximum evolutionary algebra, g j 'is the jth gene of the individual G u after evolution;
  • the optimal population individual will be output as the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index.
  • N represents N moments before the current moment.
  • the value of N can be set to a fixed value or a sliding window size of a defined time according to requirements, and the statistical result will be based on a fixed reference value or a sliding reference value Set, where the average amount of illegality in N time period is The average amount of reconsideration is Reconsideration rate For example, to consider changes in recent data, set N as the recent time period; to consider the impact of all data, set N as all time periods.
  • the health of law enforcement is affected by many factors due to different policies in different cities over time, and a time attenuation factor is added to constrain the changes brought about by various factors at different times.
  • the data includes the average amount of violations Z M , the average amount of reconsideration Z f , the average reconsideration rate Z p , the maximum amount of violations max (M N ) and the minimum amount of violations min (M N ), the maximum amount of reconsideration max (f N ) and the minimum amount of reconsideration min(f N ), the maximum reconsideration rate max(p N ), and the minimum reconsideration rate min(p N )) assign a time attenuation factor.
  • the closer to the current moment the greater the time attenuation factor at other moments, that is, the closer to the current moment.
  • each indicator we know the specific factors that affect the decline in health, so that relevant departments can make decisions. For example, when the illegal amount is too large, send an alarm signal to the public security department and the neighborhood committees.
  • the strategy is that the public security department strengthens law enforcement and neighborhood committees. Strengthen the generalization of the law; when the amount of reconsideration is too large, increase the rules and descriptions of violations, so that the citizens can understand the reasons for violations more clearly; improve the quality of law enforcement personnel and the standards for judging violations; the reconsideration rate is too large, and the scale and intensity of law enforcement by illegal personnel will be increased.
  • step S4 the illegal incremental change index, the reconsideration incremental change index, and the weighted incremental value of the reconsideration rate change index ⁇ W M *B′ M , W f * are obtained through the following formula (13) B′ f , W p *B′ p ⁇ , the weighted increment value is negatively related to law enforcement health, such as [W M *B′ M >W f *B′ f >W p *B′ p ] means The illegal incremental change index has the greatest impact on health,
  • B 'i is the incremental change law index, or index reconsideration reconsideration incremental change rate variation index increment of time t-1 to time t.
  • this application also provides a law enforcement analysis device, including a law enforcement terminal and a server.
  • the law enforcement terminal includes a law enforcement recorder and a mobile terminal with a shooting function.
  • the server includes an illegal amount obtaining unit, a reconsideration amount obtaining unit, and a reconsideration rate.
  • Acquisition unit, law enforcement change index acquisition unit and law enforcement analysis unit including:
  • the law enforcement terminal collects event pictures and sends them to the server;
  • the illegal quantity obtaining unit extracts image features and judges whether it is illegal, the sum of illegal events at one moment is regarded as the illegal quantity, and the illegal quantity at different times constitutes the illegal quantity data set;
  • the reconsideration amount obtaining unit uses the sum of the events that are judged to be illegal at one time and proposes reconsideration as the reconsideration amount, and the reconsideration amount data set constituted by the reconsideration amounts at different moments;
  • the reconsideration rate obtaining unit obtains the illegal amount data set obtained by the illegal amount obtaining unit and the reconsideration amount data set obtained by the reconsideration amount obtaining unit according to the following formula to obtain the reconsideration rate data set
  • F t is the illegal amount at time t
  • M t is the amount of reconsideration at time t
  • p t is the reconsideration rate at time t
  • the law enforcement change index obtaining unit obtains the illegal incremental change index, the reconsideration incremental change index, and the reconsideration rate change index at each moment through the illegal amount data set, the reconsideration amount data set, and the reconsideration rate data set;
  • the law enforcement analysis unit obtains law enforcement health through the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index.
  • the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index are negative for law enforcement health. Related.
  • the embodiment of the present application also proposes a computer-readable storage medium.
  • the computer-readable storage medium may be non-volatile or volatile.
  • the computer-readable storage medium includes a law enforcement analysis program. When the law enforcement analysis program is executed by the processor, the following steps are implemented:
  • Step S1 Send the event pictures collected by the law enforcement terminal to the server, extract the image features of the event pictures and determine whether they are illegal, and collect the number of illegal events at different moments to form a data set of illegal quantities, the server Collecting the number of incidents determined to be illegal and requesting reconsideration, and obtaining a reconsideration amount data set composed of reconsideration amounts at different moments, the law enforcement terminal includes a law enforcement recorder and a mobile terminal with a shooting function;
  • Step S2 Obtain the reconsideration rate data set according to the following formula through the illegal amount data set and the reconsideration amount data set uploaded to the server:
  • F t is the illegal amount at time t
  • M t is the amount of reconsideration at time t
  • p t is the reconsideration rate at time t
  • Step S3 The server obtains the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index at each moment through the illegal amount data set, the reconsideration amount data set, and the reconsideration rate data set;
  • Step S4 the server obtains law enforcement health through the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index, where the illegal incremental change index, reconsideration incremental change index, and reconsideration rate change index are negatively correlated with law enforcement health .
  • the above-mentioned law enforcement analysis methods, devices, electronic equipment, and computer-readable storage media use reconsideration amount, reconsideration rate, and illegal amount as the main evaluation indicators, and add a time attenuation factor to each indicator to improve the influence of recent events and reduce historical data The impact of this is as close to reality as possible.
  • the law enforcement analysis method further includes optimizing the police force distribution density according to law enforcement health, which specifically includes:
  • PL ij is the police distribution density of the j-th street in the i-th district
  • a ij is the construction area of the j-th street in the i-th district
  • a i is the construction area of the i-th district
  • e is the accuracy error. The higher the accuracy, the smaller the value of e;
  • PL ij is greater than The range indicates that the current actual police force distribution on the street is small, and the police force can be increased.
  • the above-mentioned law enforcement analysis methods, devices, electronic equipment and computer-readable storage media evaluate the health of law enforcement in a city or a region through the main evaluation indicators such as police distribution density, reconsideration amount, reconsideration rate, and illegal amount. They are applicable to different cities. The health of cities can be comparable.
  • the information collected by the law enforcement terminal is not limited to time pictures, and may also be text describing events.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Strategic Management (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Biophysics (AREA)
  • Tourism & Hospitality (AREA)
  • Economics (AREA)
  • Educational Administration (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • General Health & Medical Sciences (AREA)
  • Development Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Genetics & Genomics (AREA)
  • Software Systems (AREA)
  • Technology Law (AREA)
  • Mathematical Physics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • Evolutionary Computation (AREA)
  • Primary Health Care (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Biomedical Technology (AREA)
  • Artificial Intelligence (AREA)
  • Physiology (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Medical Treatment And Welfare Office Work (AREA)

Abstract

本申请涉及数据挖掘，提供一种执法分析方法，包括：将执法终端采集的事件图片发送到服务器，对事件图片提取图像特征并对是否违法进行判别，获得不同时刻的违法量构成违法量数据集，还采集判定为违法并提出复议的事件，获得复议量数据集；通过上传至服务器的违法量数据集和复议量数据集获得复议率数据集；服务器通过上述数据集分别获得各时刻的违法增量变化指数、复议增量变化指数和复议率变化指数；服务器通过违法增量变化指数、复议增量变化指数和复议率变化指数获得执法健康度，违法增量变化指数、复议增量变化指数和复议率变化指数与执法健康度负相关。本申请还提供一种装置、电子设备及存储介质。本申请客观分析执法是否文明是否健康。

Description

执法分析方法、装置、电子设备及存储介质
本申请要求于2019年8月14日提交中国专利局、申请号为201910748995.2，发明名称为“执法分析方法、装置及介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及数据挖掘技术领域，更为具体地，涉及一种执法分析方法、装置、电子设备及存储介质。
背景技术
文明执法，就是在行政执法中树立以人为本、依法行政、执政为民的理念，充分尊重行政执法相对人的权益，严格遵循法律规定的执法程序，坚持教育与处罚相结合，管理与服务相结合，不断提高行政执法效能，为建设和谐社会和法制社会提供保障。
现有技术中，通过违法事件、复议事件的图片、影像和数量等人为主观分析文明执法程度，发明人意识到现有技术没有考虑到随时间变化执法情况的变化，不存在对文明执法的客观评价方法，无法判断执法是否健康，健康执法关系到社会的各行行业，现有技术中没有对执法健康度进行分析的方法。另外，在大数据时代，数据挖掘依靠认为客观分析，准确率和工作效率低。
发明内容
鉴于上述问题，本申请的目的是提供一种客观分析执法是否文明是否健康的执法分析方法、装置、电子设备及存储介质。
为了实现上述目的，本申请提供一种执法分析方法，包括：
步骤S1，将执法终端采集的事件图片发送到服务器，对所述事件图片提取图像特征并对是否违法进行判别，采集不同时刻的被判别为违法的事件的数量，构成违法量数据集，所述服务器采集判定为违法并提出复议的事件的数量，获得不同时刻的复议量构成的复议量数据集，所述执法终端包括执法记录仪和具有拍摄功能的移动终端；
步骤S2，通过上传至服务器的违法量数据集和复议量数据集根据下式获得复议率数据集：
Figure PCTCN2020105341-appb-000001
其中，F t为t时刻的违法量，M t为t时刻的复议量，p t为t时刻的复议率；
步骤S3，服务器通过违法量数据集、复议量数据集和复议率数据集获得各时刻的违法增量变化指数、复议增量变化指数和复议率变化指数；
步骤S4，服务器通过违法增量变化指数、复议增量变化指数和复议率变化指数获得执法健康度，所述违法增量变化指数、复议增量变化指数和复议率变化指数与执法健康度负相关。
此外，为了实现上述目的，本申请还提供一种执法分析装置，包括执法终端和服务器，所述执法终端包括执法记录仪和具有拍摄功能的移动终端，所述服务器包括违法量获得单元、复议量获得单元、复议率获得单元、执法变化指数获得单元和执法分析单元，其中：
所述执法终端采集事件图片并发送到服务器；
所述违法量获得单元提取图像特征并对是否违法进行判别，一个时刻违法事件的总和作为违法量，不同时刻的违法量构成违法量数据集；
所述复议量获得单元将一个时刻判定为违法并提出复议的事件的总和作为复议量，不同时刻的复议量构成的复议量数据集；
所述复议率获得单元通过违法量获得单元获得的违法量数据集和复议量获得单元获 得的复议量数据集根据下式获得复议率数据集
Figure PCTCN2020105341-appb-000002
其中，F t为t时刻的违法量，M t为t时刻的复议量，p t为t时刻的复议率；
所述执法变化指数获得单元通过违法量数据集、复议量数据集和复议率数据集分别获得各时刻的违法增量变化指数、复议增量变化指数和复议率变化指数；
所述执法分析单元通过违法增量变化指数、复议增量变化指数和复议率变化指数获得执法健康度，所述违法增量变化指数、复议增量变化指数和复议率变化指数与执法健康度负相关。
此外，为了实现上述目的，本申请还提供一种电子设备，包括存储器和处理器，所述存储器中存储有执法分析程序，所述执法分析程序被所述处理器执行时实现上述的执法分析方法的步骤。
此外，为了实现上述目的，本申请还提供一种计算机可读存储介质，所述计算机可读存储介质中包括执法分析程序，所述执法分析程序被处理器执行时，实现上述的执法分析方法的步骤。
本申请所述执法分析方法、装置、电子设备及计算机可读存储介质可以通过量化的数值直观的对执法健康度有一个认识，违法增量变化指数、复议增量变化指数和复议率变化指数都加入了增量信息，不仅用执法数量还使用了执法指数变化方向，正向或负向信息，客观分析执法健康度，并可以通过各个指标的影响因素知道具体影响健康度下降的因素，以便相关部门决策。
附图说明
图1是本申请执法分析方法较佳实施例的应用环境示意图；
图2是图1中执法分析程序较佳实施例的模块示意图；
图3是本申请执法分析方法较佳实施例的流程图。
具体实施方式
应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
以下将结合附图对本申请的具体实施例进行详细描述。
本申请提供一种执法分析方法，应用于一种电子装置1。参照图1所示，为本申请执法分析方法较佳实施例的应用环境示意图。
在本实施例中，电子装置1可以是服务器、手机、平板电脑、便携计算机、桌上型计算机等具有运算功能的终端客户端。
存储器11包括至少一种类型的可读存储介质。所述至少一种类型的可读存储介质可为如闪存、硬盘、多媒体卡、卡型存储器等的非易失性存储介质。在一些实施例中，所述可读存储介质可以是所述电子装置1的内部存储单元，例如该电子装置1的硬盘。在另一些实施例中，所述可读存储介质也可以是所述电子装置1的外部存储器，例如所述电子装置1上配备的插接式硬盘，智能存储卡(Smart Media Card，SMC)，安全数字(Secure Digital，SD)卡，闪存卡(Flash Card)等。
在本实施例中，所述存储器11的可读存储介质通常用于存储安装于所述电子装置1的执法分析程序10等。所述存储器11还可以用于暂时地存储已经输出或者将要输出的数据。
处理器12在一些实施例中可以是一中央处理器(Central Processing Unit，CPU)，微处理器或其他数据处理芯片，用于运行存储器11中存储的程序代码或处理数据，例如执行执法分析程序10等。
网络接口13可选地可以包括标准的有线接口、无线接口(如WI-FI接口)，通常用于在 该电子装置1与其他电子客户端之间建立通信连接。
通信总线14用于实现这些组件之间的连接通信。
图1仅示出了具有组件11-14的电子装置1，但是应理解的是，并不要求实施所有示出的组件，可以替代的实施更多或者更少的组件。
可选地，该电子装置1还可以包括用户接口，用户接口可以包括输入单元比如键盘(Keyboard)、语音输入装置比如麦克风(microphone)等具有语音识别功能的客户端、语音输出装置比如音响、耳机等，可选地用户接口还可以包括标准的有线接口、无线接口。
可选地，该电子装置1还可以包括显示器，显示器也可以称为显示屏或显示单元。
在一些实施例中可以是LED显示器、液晶显示器、触控式液晶显示器以及有机发光二极管(Organic Light-Emitting Diode，OLED)触摸器等。显示器用于显示在电子装置1中处理的信息以及用于显示可视化的用户界面。
可选地，该电子装置1还包括触摸传感器。所述触摸传感器所提供的供用户进行触摸操作的区域称为触控区域。此外，这里所述的触摸传感器可以为电阻式触摸传感器、电容式触摸传感器等。而且，所述触摸传感器不仅包括接触式的触摸传感器，也可包括接近式的触摸传感器等。此外，所述触摸传感器可以为单个传感器，也可以为例如阵列布置的多个传感器。
可选地，该电子装置1还可以包括逻辑门电路，传感器、音频电路等等，在此不再赘述。
在图1所示的装置实施例中，作为一种计算机存储介质的存储器11中可以包括操作系统以及执法分析程序10；处理器12执行存储器11中存储的执法分析程序10时实现如下步骤：
步骤S1，将执法终端采集的事件图片发送到服务器，所对所述事件图片提取图像特征并对是否违法进行判别，采集不同时刻的被判别为违法的事件的数量，构成违法量数据集，所述服务器采集判定为违法并提出复议的事件的数量，获得不同时刻的复议量构成的复议量数据集，所述执法终端包括执法记录仪和具有拍摄功能的移动终端(手机、相机等)，例如，执法人员在实际执法中将违法事件通过执法记录仪或手机通过网络上传至服务器构成违法量数据集，违法数据和记录将通过邮件，短信，电话等方式通知违法人员，违法人员在违法存在争议的情况下可通过执法记录官网在在线提出复议，或通过人工办事窗口提出复议，复议数据将上传至服务器，构成复议量数据集；
步骤S2，通过上传至服务器的违法量数据集和复议量数据集根据下式(1)获得复议率数据集
Figure PCTCN2020105341-appb-000003
其中，F t为t时刻的违法量，M t为t时刻的复议量，p t为t时刻的复议率；
步骤S3，服务器通过违法量数据集、复议量数据集和复议量数据集根据下式(2)-(4)分别获得各时刻的违法增量变化指数、复议增量变化指数和复议率变化指数
Figure PCTCN2020105341-appb-000004
Figure PCTCN2020105341-appb-000005
Figure PCTCN2020105341-appb-000006
其中，N为当前时刻前设定时间段，Z M为所述时间段的平均违法量、Z f为所述时间段的平均复议量、Z p为所述时间段的平均复议率、max(M N)为所述时间段的最大违法量、min(M N)为所述时间段的最小违法量、max(f N)为所述时间段的最大复议量、min(f N)为所述时间段的最小复议量、max(p N)为所述时间段的最大复议率、min(p N)为所述时间段的最小复议率、B M为违法增量变化指数、B f为复议增量变化指数、B p为复议率变化指数；
步骤S4，服务器通过违法增量变化指数、复议增量变化指数和复议率变化指数获得执法健康度，所述违法增量变化指数、复议增量变化指数和复议率变化指数与执法健康度负相关，也就是说，所述违法增量变化指数、复议增量变化指数和复议率变化指数越大，执法健康度越低，其中，所述执法健康度是衡量行政主体执法的健康程度(文明执法程度)的一个量化指标。
优选地，处理器12执行存储器11中存储的执法分析程序10时还实现如下步骤：在步骤S3和步骤S4之间还包括：对违法增量变化指数、复议增量变化指数和复议率变化指数根据下式(5)进行加权组合，得到综合指数
Figure PCTCN2020105341-appb-000007
其中，Z为综合指数，W M、W f和W p分别为B M、B f和B p的权重,W M+W f+W p＝1。
此时，在步骤S4中，综合指数与执法健康度正相关，也就是说，随着综合指数越大，执法健康度越高，例如，综合指数取值范围[-1，1]，经过数值正负号的转换，Z大于0表示执法健康度好，Z小于0表示执法健康度低，Z等与0表示执法健康度正常。
在其他实施例中，所述执法分析程序10还可以被分割为一个或者多个模块，一个或者多个模块被存储于存储器11中，并由处理器12执行，以完成本申请。本申请所称的模块是指能够完成特定功能的一系列计算机程序指令段。参照图2所示，为图1中执法分析程序10较佳实施例的功能模块图。所述执法分析程序10可以被分割为采集模块110和服务器120，服务器120包括违法量获得单元121、复议量获得单元122、复议率获得单元123、执法变化指数获得单元124和执法分析单元125，其中，所述采集模块110将执法终端采集的事件图片发送到服务器120；违法量获得单元121提取图像特征并对是否违法进行判别，一个时刻违法事件的总和作为违法量，不同时刻的违法量构成违法量数据集；所述复议量获得单元122将一个时刻判定为违法并提出复议的事件的总和作为复议量，不同时刻的复议量构成的复议量数据集；所述复议率获得单元123通过违法量获得单元121获得的违法量数据集和复议量获得单元122获得的复议量数据集获得复议率数据集；所述执法变化指数获得单元124通过违法量数据集、复议量数据集和复议率数据集分别获得各时刻的违法增量变化指数、复议增量变化指数和复议率变化指数；所述执法分析单元126通过违法增量变化指数、复议增量变化指数和复议率变化指数获得执法健康度。
优选地，服务器120还包括综合指数获得单元125对违法增量变化指数、复议增量变化指数和复议率变化指数根据下式进行加权组合，得到综合指数，此时，执法分析单元根据综合指数分析执法健康度，随着综合指数越大，执法健康度越高。
此外，本申请还提供一种执法分析方法。参照图3所示，为本申请执法分析方法较佳实施例的流程图。该方法可以由一个装置执行，该装置可以由软件和/或硬件实现。
在本实施例中，执法分析方法包括：
步骤S1，将执法终端采集的事件图片发送到服务器对所述事件图片提取图像特征并对是否违法进行判别，采集不同时刻的被判别为违法的事件的数量构成违法量数据集，所述服务器采集判定为违法并提出复议的事件的数量，获得不同时刻的复议量构成的复议量数据集，所述执法终端包括执法记录仪和具有拍摄功能的移动终端(手机、相机等)；
步骤S2，通过上传至服务器的违法量数据集和复议量数据集根据下式(1)获得复议率数据集
Figure PCTCN2020105341-appb-000008
其中，F t为t时刻的违法量，M t为t时刻的复议量，p t为t时刻的复议率；
步骤S3，服务器通过违法量数据集、复议量数据集和复议量数据集根据下式(2)-(4)分别获得各时刻的违法增量变化指数、复议增量变化指数和复议率变化指数
Figure PCTCN2020105341-appb-000009
Figure PCTCN2020105341-appb-000010
Figure PCTCN2020105341-appb-000011
其中，N为当前时刻前设定时间段，Z M为所述时间段的平均违法量、Z f为所述时间段的平均复议量、Z p为所述时间段的平均复议率、max(M N)为所述时间段的最大违法量、min(M N)为所述时间段的最小违法量、max(f N)为所述时间段的最大复议量、min(f N)为所述时间段的最小复议量、max(p N)为所述时间段的最大复议率、min(p N)为所述时间段的最小复议率；
步骤S4，服务器通过违法增量变化指数、复议增量变化指数和复议率变化指数获得执法健康度，所述违法增量变化指数、复议增量变化指数和复议率变化指数与执法健康度负相关，也就是说，所述违法增量变化指数、复议增量变化指数和复议率变化指数越大，执法健康度越低。
在一个可选实施例中，在步骤S1中，所述违法量的获得方法包括：
将执法人员执法记录仪及执法人员和群众的具有拍摄功能的移动终端拍摄的可能违法的事件的图片上传至图像分析服务器，抽取图像特征，例如，执法人员执法记录仪、执法人员手机拍摄图片和群众拍摄图片通过公众号和官网上传至图像分析服务器，图像分析服务器由高性的人工智能图像识别技术组成，通过图像识别算法抽取图像特征；
将上述图像特征输入卷积神经网络，获得所述图片对应的事件的违法概率，例如，采用VGGNet、AlexNet或GoogleNet等卷积神经网络，使用卷积深度学习的方法训练违法图片的分类模型。分类模型训练需要提前标注一批违法照片，混入其他照片，通过模型训练分类器，分类照片是否为违法照片。分类模型预测输入照片，输出照片违法概率或是否违法状态，例如，卷积神经网络输入图片大小224x224像素的RGB图片，经过多层3D卷积和最大池化操作，再添加神经网络的全连接层，最终通过将图片映射为1000维的特征向量，特征向量作为二分类器(逻辑回归、决策树、随机森林、支持向量机)的输入训练分类器，分类器可以由神经网络全连接到二维特征向量上，使用softmax函数作为目标函数进行映射，输出最终结果违法概率值，取值在0-1的小数，数值越大违法可能性越高，违 法事实约肯定，也可以使用机器学习算法的logist，svm，Tree等模型进行二分类，输出违法概率；
将违法概率超过第一设定阈值的事件作为违法事件；
对不同时刻的违法事件的数量进行统计，进而获得不同时刻的违法量，优选地，将图像和对应的违法信息和违法状态记录到为服务器的数据库，构成违法量数据集，所述违法信息包括违法地点、违法人员、执法人员、群众举报人、联系方式、违法照片、是否违法状态等信息。
在步骤S1中，所述复议量数据集和违法量数据集的获得方法包括：
获得提出复议的违法事件的数量，从而得到复议量数据集，例如，将违法数据集通过短信，邮件，微信，公众号的方式发送违法信息给违法人，违法人在对违法信息存在争议时通过，公众号，人工受理窗口，政府官网等渠道进行复议；
将提出复议，还没有获得复议结果的违法事件的违法概率、违法人违法记录次数、违法严重程度和违法人信用等级进行加权求和，获得所述违法事件的违法置信度；
统计违法置信度不大于第二设定阈值的违法事件及复议成功的违法事件；
在违法量数据集中删除所述置信度不大于第二设定阈值的违法事件和复议成功的违法事件，对违法量数据集进行更新，也就是说，判别该违法记录复议成功或失败，复议成功将撤销违法人的违法记录和违法处罚，复议失败将保持违法记录的违法状态，将复议的流程记录到数据库中，其中，所有提起复议的数据就构成复议量数据集。
在一个可选实施例中，在步骤S3和步骤S4之间还包括：对违法增量变化指数、复议增量变化指数和复议率变化指数根据下式(5)进行加权组合，得到综合指数
Figure PCTCN2020105341-appb-000012
其中，Z为综合指数，W M、W f和W p分别为B M、B f和B p的权重，W M+W f+W p＝1，
此时，在步骤S4中，综合指数与执法健康度正相关，也就是说，随着综合指数越大，执法健康度越高。
优选地，步骤S4包括：
通过映射函数对综合指数进行数值的映射，根据下式(6)获得执法健康度
E＝f(Z)   (6)
其中，
Figure PCTCN2020105341-appb-000013
x＝[-1，1]，f(x)＝[a，b]，[a，b]为映射区间，E为执法健康度，执法健康度与执法健康程度正相关。
优选地，在步骤S3之后，所述方法还包括：
通过正态分布对违法增量变化指数、复议增量变化指数、复议率变化指数或/和综合指数进行异常值检测，包括：
通过下式下式(7)和(8)分别获得各个指数的均值和标准差
Figure PCTCN2020105341-appb-000014
Figure PCTCN2020105341-appb-000015
其中，X i为违法增量变化指数、复议增量变化指数、复议率变化指数和综合指数中一个指数第i时刻的值，μ为所述指数的平均值，σ为所述指数的标准差；
通过下式(9)判断各时刻的指数的值得真和假
Figure PCTCN2020105341-appb-000016
其中，为违法增量变化指数、复议增量变化指数、复议率变化指数和综合指数中一个 指数经判别后的结果，为真值(正常值)或为假值(异常值)，可以采用二值化表示真假结果。
上述异常值检测剔除特别大的不符合常规的值，防止因为某几个异常值导致整体的计算结果偏小，集中在某个很小的数据区间，不便于后续计算，并且很难体现结果之间的差异。
在本申请的一个可选实施例中，在步骤S3和步骤S4之间还包括：采用遗传算法对违法增量变化指数、复议增量变化指数和复议率变化指数的权重进行优化的步骤，所述步骤包括：
采集样本，所述样本包括违法量数据集及其对应的复议量数据集；
通过专家知识库获得多个样本的健康度；
设群体规模为P，随机生成P个个体的初始种群，G＝(G 1，G 2，…，G p) T，挑选[0，1]内的随机实数组成长度为3的非全零实数向量，赋予种群中个体G i＝(g 1，g 2，g 3)，i＝1，2，…，P，g 1为个体G i中的第1个基因；
将每一个个体的各个基因分别作为违法增量变化指数、复议增量变化指数和复议率变化指数的权重，根据属于每一个话题类的样本的综合指数，得到每一个个体的适应度，其中：
Figure PCTCN2020105341-appb-000017
其中，
Figure PCTCN2020105341-appb-000018
为初始种群G中个体G i的适应度，m为样本数，K为样本索引，E K为综合指数，E＇ K为通过专家知识库获得样本的健康度；
采用轮盘赌算子，基于适应度比例的选择策略对初始种群中的个体进行选择，得到选出个体G u；
采用单点交叉算子，对选出个体进行交叉更新进行交叉更新，将更新后每个基因的最大值，作为所述基因的上界，将更新后的每个基因的最小值作为所述基因的下界；
对经过交叉更新的选出个体进行变异操作，得到变异后的个体，代入个体评价子单元，对初始种群进行进化，其中：
Figure PCTCN2020105341-appb-000019
Figure PCTCN2020105341-appb-000020
其中，g j为选出个体G u的第j个基因，g jmax和g jmin是基因g j的上界和下界，r p为在选出个体G u时第P次生成的伪随机数，iter now是当前的进化代数，iter max是设置的最大进化代数，g j＇为进化后的个体G u的第j个基因；
判断遗传算法是否满足算法结束条件，其中，所述算法结束条件包括当前进化代数大于所设最大进化代数或连续多次进化时个体适应度值变化小于所设目标值；
将满足算法结束条件，则输出最优的种群个体，作为违法增量变化指数、复议增量变化指数和复议率变化指数。
在上述各实施例中，在步骤S3中，N表示当前时刻前N个时刻，N的值可以根据需求设定固定值或者定义时间的滑动窗口大小，统计结果将按照固定参照值或者滑动参照值设定，其中，N时间段的平均违法量为
Figure PCTCN2020105341-appb-000021
平均复议量为
Figure PCTCN2020105341-appb-000022
复议率为
Figure PCTCN2020105341-appb-000023
例如，要考虑近期数据的变化情况，则设定N为近期时间段；要考虑所有数据的影响，则设定N为所有时间段。
优选地，因不同城市间政策不同随时间变化执法健康度收到诸多因素影响，加入时间衰减因子来约束因不同时间的各种因素带来的变化，具体地：对不同时刻的数据(所述数据包括平均违法量Z M、平均复议量Z f、平均复议率Z p、最大违法量max(M N)和最小违法量min(M N)、最大复议量max(f N)和最小复议量min(f N)、最大复议率max(p N)和最小复议率min(p N))赋予时间衰减因子，越接近当前时刻的其他时刻的时间衰减因子越大，也就是说，越接近当前时刻的其他时刻的权重越大。
优选地，通过各个指标的影响因素知道具体影响健康度下降的因素，以便相关部门决策，例如：违法量过大的时候，发送报警信号给公安部门和各居委会，策略是公安部门加强执法和居委会加强普法；复议量过大时，增加违法违规的规则说明，让市民更清楚了解违法的原因；提高执法人员的素质和违法判断的标准；复议率过大，增加违法人员执法的尺度和力度。
进一步，优选地，在步骤S4中，通过下式(13)，获得违法增量变化指数、复议增量变化指数和复议率变化指数的加权增量值{W M*B′ M，W f*B′ f，W p*B′ p}，所述加权增量值与执法健康度负相关，如[W M*B′ M＞W f*B′ f＞W p*B′ p]则表示违法增量变化指数对健康度影响最大，
Figure PCTCN2020105341-appb-000024
其中，B′ i为违法增量变化指数、复议增量变化指数或复议率变化指数在t-1时刻到t时刻的增量。
此外，本申请还提供一种执法分析装置，包括执法终端和服务器，所述执法终端包括执法记录仪和具有拍摄功能的移动终端，所述服务器包括违法量获得单元、复议量获得单元、复议率获得单元、执法变化指数获得单元和执法分析单元，其中：
所述执法终端采集事件图片并发送到服务器；
所述违法量获得单元提取图像特征并对是否违法进行判别，一个时刻违法事件的总和作为违法量，不同时刻的违法量构成违法量数据集；
所述复议量获得单元将一个时刻判定为违法并提出复议的事件的总和作为复议量，不同时刻的复议量构成的复议量数据集；
所述复议率获得单元通过违法量获得单元获得的违法量数据集和复议量获得单元获得的复议量数据集根据下式获得复议率数据集
Figure PCTCN2020105341-appb-000025
其中，F t为t时刻的违法量，M t为t时刻的复议量，p t为t时刻的复议率；
所述执法变化指数获得单元通过违法量数据集、复议量数据集和复议率数据集分别获得各时刻的违法增量变化指数、复议增量变化指数和复议率变化指数；
所述执法分析单元通过违法增量变化指数、复议增量变化指数和复议率变化指数获得执法健康度，所述违法增量变化指数、复议增量变化指数和复议率变化指数与执法健康度负相关。
此外，本申请实施例还提出一种计算机可读存储介质，所述计算机可读存储介质可以是非易失性，也可以是易失性，所述计算机可读存储介质中包括执法分析程序，所述执法分析程序被处理器执行时实现如下步骤：
步骤S1，将执法终端采集的事件图片发送到服务器，对所述事件图片提取图像特征并对是否违法进行判别，采集不同时刻的被判别为违法的事件的数量构成违法量数据集，所述服务器采集判定为违法并提出复议的事件的数量，获得不同时刻的复议量构成的复议量数据集，所述执法终端包括执法记录仪和具有拍摄功能的移动终端；
步骤S2，通过上传至服务器的违法量数据集和复议量数据集根据下式获得复议率数据 集：
Figure PCTCN2020105341-appb-000026
其中，F t为t时刻的违法量，M t为t时刻的复议量，p t为t时刻的复议率；
步骤S3，服务器通过违法量数据集、复议量数据集和复议率数据集获得各时刻的违法增量变化指数、复议增量变化指数和复议率变化指数；
步骤S4，服务器通过违法增量变化指数、复议增量变化指数和复议率变化指数获得执法健康度，所述违法增量变化指数、复议增量变化指数和复议率变化指数与执法健康度负相关。
本申请之执法分析装置、计算机可读存储介质的具体实施方式与上述执法分析方法、电子装置的具体实施方式大致相同，在此不再赘述。
上述执法分析方法、装置、电子设备及计算机可读存储介质通过复议量，复议率，违法量作为主要评估指标，再各项指标上加入时间衰减因子，提高近期事件发展规律的影响，降低历史数据的影响，尽可能贴近现实。
在上述执法分析方法、装置、电子设备及计算机可读存储介质的各个实施例中，优选地，还包括根据执法健康度优化警力分配密度，具体地包括：
根据下式(14)构建警力分配密度模型
Figure PCTCN2020105341-appb-000027
其中，PL ij为第i区第j个街道的警力分布密度；
判断PL ij在
Figure PCTCN2020105341-appb-000028
范围内，其中，a ij为第i区第j个街道的建筑面积，a i为第i区的建筑面积，e是精度误差，要求精度越高e值越小；
如果PL ij在
Figure PCTCN2020105341-appb-000029
范围内，说明所述街道目前的实际的警力分布是合理的；
如果PL ij小于
Figure PCTCN2020105341-appb-000030
范围，说明所述街道目前的实际的警力分布多，可以减少警力；
如果PL ij大于
Figure PCTCN2020105341-appb-000031
范围，说明所述街道目前的实际的警力分布少，可以增加警力。
上述执法分析方法、装置、电子设备及计算机可读存储介质通过警力分配密度，复议量，复议率，违法量等主要评估指标评价一个城市，或一个地区执法的健康度，适用于不同城市，不同城市之间的健康度可具有一定的可比性。
在本申请的上述各实施例中，执法终端采集的信息不限于时间图片，还可以是描述事件的文字。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、装置、物品或者方法不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、装置、物品或者方法所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、装置、物品或者方法中还存在另外的相同要素。
上述本申请实施例序号仅仅为了描述，不代表实施例的优劣。通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在如上所述的一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端客户端(可以是手机，计算机，服务器，或者网络客户端等)执行本申请各个实施例所述的方法。
以上仅为本申请的优选实施例，并非因此限制本申请的专利范围，凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本申请的专利保护范围内。

Claims (20)

  1. 一种执法分析方法，其中，包括：
    步骤S1，将执法终端采集的事件图片发送到服务器，对所述事件图片提取图像特征并对是否违法进行判别，采集不同时刻的被判别为违法的事件的数量，构成违法量数据集，所述服务器采集判定为违法并提出复议的事件的数量，获得不同时刻的复议量构成的复议量数据集，所述执法终端包括执法记录仪和具有拍摄功能的移动终端；
    步骤S2，通过上传至服务器的违法量数据集和复议量数据集根据下式获得复议率数据集：
    Figure PCTCN2020105341-appb-100001
    其中，F t为t时刻的违法量，M t为t时刻的复议量，p t为t时刻的复议率；
    步骤S3，服务器通过违法量数据集、复议量数据集和复议率数据集获得各时刻的违法增量变化指数、复议增量变化指数和复议率变化指数；
    步骤S4，服务器通过违法增量变化指数、复议增量变化指数和复议率变化指数获得执法健康度，所述违法增量变化指数、复议增量变化指数和复议率变化指数与执法健康度负相关。
  2. 根据权利要求1所述的执法分析方法，其中，在步骤S3和步骤S4之间还包括：
    对违法增量变化指数、复议增量变化指数和复议率变化指数根据下式进行加权组合，得到综合指数：
    Figure PCTCN2020105341-appb-100002
    其中，Z为综合指数，W M、W f和W p分别为违法增量变化指数B M、复议增量变化指数B f和复议率变化指数B p的权重,W M+W f+W p＝1，
    其中，在步骤S4中，综合指数与执法健康度正相关。
  3. 根据权利要求2所述的执法分析方法，其中，步骤S4包括：
    通过映射函数对综合指数进行数值的映射，获得执法健康度
    E＝f(Z)
    其中，
    Figure PCTCN2020105341-appb-100003
    x＝[-1，1]，f(x)＝[a，b]，[a，b]为映射区间，E为执法健康度，执法健康度与执法健康程度正相关。
  4. 根据权利要求2所述的执法分析方法，其中，在步骤S4中，通过下式获得违法增量变化指数、复议增量变化指数和复议率变化指数的加权增量值{W M*B′ M，W f*B′ f，W p*B′ p}，所述加权增量值与执法健康度负相关
    Figure PCTCN2020105341-appb-100004
    其中，B′ i为违法增量变化指数、复议增量变化指数或复议率变化指数在t-1时刻到t时刻的增量。
  5. 根据权利要求2所述的执法分析方法，其中，在步骤S3和步骤S4之间还包括：采用遗传算法对违法增量变化指数、复议增量变化指数和复议率变化指数的权重进行优化的步骤，所述步骤包括：
    采集样本，所述样本包括违法量数据集及其对应的复议量数据集；
    通过专家知识库获得多个样本的健康度；
    设群体规模为P，随机生成P个个体的初始种群，G＝(G 1，G 2，…，G p) T，挑选[0，1]内的随机实数组成长度为3的非全零实数向量，赋予种群中个体G i＝(g 1，g 2，g 3)，i＝ 1，2，…，P，g 1为个体G i中的第1个基因；
    将每一个个体的各个基因分别作为违法增量变化指数、复议增量变化指数和复议率变化指数的权重，根据属于每一个话题类的样本的综合指数，得到每一个个体的适应度，其中：
    Figure PCTCN2020105341-appb-100005
    其中，
    Figure PCTCN2020105341-appb-100006
    为初始种群G中个体G i的适应度，m为样本数，K为样本索引，E K为综合指数，E＇ K为通过专家知识库获得样本的健康度；
    采用轮盘赌算子，基于适应度比例的选择策略对初始种群中的个体进行选择，得到选出个体G u；
    采用单点交叉算子，对选出个体进行交叉更新进行交叉更新，将更新后每个基因的最大值，作为所述基因的上界，将更新后的每个基因的最小值作为所述基因的下界；
    对经过交叉更新的选出个体进行变异操作，得到变异后的个体，代入个体评价子单元，对初始种群进行进化，其中：
    Figure PCTCN2020105341-appb-100007
    Figure PCTCN2020105341-appb-100008
    其中，g j为选出个体G u的第j个基因，g jmax和g jmin是基因g j的上界和下界，r p为在选出个体G u时第P次生成的伪随机数，iter now是当前的进化代数，iter max是设置的最大进化代数，g j＇为进化后的个体G u的第j个基因；
    判断遗传算法是否满足算法结束条件，其中，所述算法结束条件包括当前进化代数大于所设最大进化代数或连续多次进化时个体适应度值变化小于所设目标值；
    将满足算法结束条件，则输出最优的种群个体，作为违法增量变化指数、复议增量变化指数和复议率变化指数。
  6. 根据权利要求1所述的执法分析方法，其中，在步骤S1中，所述违法量的获得方法包括：
    将执法人员执法记录仪及执法人员和群众的具有拍摄功能的移动终端拍摄的可能违法的事件的图片上传至图像分析服务器，抽取图像特征；
    将上述图像特征输入卷积神经网络，获得所述图片对应的事件的违法概率；
    将违法概率超过第一设定阈值的事件作为违法事件；
    对不同时刻的违法事件的数量进行统计，进而获得不同时刻的违法量。
  7. 根据权利要求6所述的执法分析方法，其中，在步骤S1中，所述复议量数据集和违法量数据集的获得方法包括：
    获得提出复议的违法事件的数量，从而得到复议量数据集；
    将提出复议，还没有获得复议结果的违法事件的违法概率、违法人违法记录次数、违法严重程度和违法人信用等级进行加权求和，获得所述违法事件的违法置信度；
    统计违法置信度不大于第二设定阈值的违法事件及复议成功的违法事件；
    在违法量数据集中删除所述置信度不大于第二设定阈值的违法事件和复议成功的违法事件，对违法量数据集进行更新。
  8. 根据权利要求1所述的执法分析方法，其中，在步骤S3之后，所述方法还包括：
    通过正态分布对违法增量变化指数、复议增量变化指数、复议率变化指数或/和综合指数进行异常值检测。
  9. 一种执法分析装置，其中，包括执法终端和服务器，所述执法终端包括执法记录仪和具有拍摄功能的移动终端，所述服务器包括违法量获得单元、复议量获得单元、复议率获得单元、执法变化指数获得单元和执法分析单元，其中：
    所述执法终端采集事件图片并发送到服务器；
    所述违法量获得单元提取图像特征并对是否违法进行判别，一个时刻违法事件的总和作为违法量，不同时刻的违法量构成违法量数据集；
    所述复议量获得单元将一个时刻判定为违法并提出复议的事件的总和作为复议量，不同时刻的复议量构成的复议量数据集；
    所述复议率获得单元通过违法量获得单元获得的违法量数据集和复议量获得单元获得的复议量数据集根据下式获得复议率数据集
    Figure PCTCN2020105341-appb-100009
    其中，F t为t时刻的违法量，M t为t时刻的复议量，p t为t时刻的复议率；
    所述执法变化指数获得单元通过违法量数据集、复议量数据集和复议率数据集分别获得各时刻的违法增量变化指数、复议增量变化指数和复议率变化指数；
    所述执法分析单元通过违法增量变化指数、复议增量变化指数和复议率变化指数获得执法健康度，所述违法增量变化指数、复议增量变化指数和复议率变化指数与执法健康度负相关。
  10. 一种电子设备，其中，包括存储器和处理器，所述存储器中存储有执法分析程序，所述执法分析程序被所述处理器执行时实现如下步骤：
    步骤S1，将执法终端采集的事件图片发送到服务器，对所述事件图片提取图像特征并对是否违法进行判别，采集不同时刻的被判别为违法的事件的数量，构成违法量数据集，所述服务器采集判定为违法并提出复议的事件的数量，获得不同时刻的复议量构成的复议量数据集，所述执法终端包括执法记录仪和具有拍摄功能的移动终端；
    步骤S2，通过上传至服务器的违法量数据集和复议量数据集根据下式获得复议率数据集：
    Figure PCTCN2020105341-appb-100010
    其中，F t为t时刻的违法量，M t为t时刻的复议量，p t为t时刻的复议率；
    步骤S3，服务器通过违法量数据集、复议量数据集和复议率数据集获得各时刻的违法增量变化指数、复议增量变化指数和复议率变化指数；
    步骤S4，服务器通过违法增量变化指数、复议增量变化指数和复议率变化指数获得执法健康度，所述违法增量变化指数、复议增量变化指数和复议率变化指数与执法健康度负相关。
  11. 根据权利要求10所述的电子设备，其中，所述执法分析程序被所述处理器执行时实现的步骤S3和步骤S4之间还包括：
    对违法增量变化指数、复议增量变化指数和复议率变化指数根据下式进行加权组合，得到综合指数：
    Figure PCTCN2020105341-appb-100011
    其中，Z为综合指数，W M、W f和W p分别为违法增量变化指数B M、复议增量变化指数B f和复议率变化指数B p的权重,W M+W f+W p＝1，
    其中，在步骤S4中，综合指数与执法健康度正相关。
  12. 根据权利要求11所述的电子设备，其中，所述执法分析程序被所述处理器执行时实现的步骤S4包括：
    通过映射函数对综合指数进行数值的映射，获得执法健康度
    E＝f(Z)
    其中，
    Figure PCTCN2020105341-appb-100012
    x＝[-1，1]，f(x)＝[a，b]，[a，b]为映射区间，E为执法健康度，执法健康度与执法健康程度正相关。
  13. 根据权利要求10所述的电子设备，其中，所述执法分析程序被所述处理器执行时实现的步骤S1包括：
    将执法人员执法记录仪及执法人员和群众的具有拍摄功能的移动终端拍摄的可能违法的事件的图片上传至图像分析服务器，抽取图像特征；
    将上述图像特征输入卷积神经网络，获得所述图片对应的事件的违法概率；
    将违法概率超过第一设定阈值的事件作为违法事件；
    对不同时刻的违法事件的数量进行统计，进而获得不同时刻的违法量。
  14. 根据权利要求13所述的电子设备，其中，所述执法分析程序被所述处理器执行时实现的步骤S1还包括：
    获得提出复议的违法事件的数量，从而得到复议量数据集；
    将提出复议，还没有获得复议结果的违法事件的违法概率、违法人违法记录次数、违法严重程度和违法人信用等级进行加权求和，获得所述违法事件的违法置信度；
    统计违法置信度不大于第二设定阈值的违法事件及复议成功的违法事件；
    在违法量数据集中删除所述置信度不大于第二设定阈值的违法事件和复议成功的违法事件，对违法量数据集进行更新。
  15. 根据权利要求10所述的电子设备，其中，所述执法分析程序被所述处理器执行时实现的步骤S3之后还包括：
    通过正态分布对违法增量变化指数、复议增量变化指数、复议率变化指数或/和综合指数进行异常值检测。
  16. 一种计算机可读存储介质，其中，所述计算机可读存储介质中包括有执法分析程序，所述执法分析程序被处理器执行时，实现如权利要求1至8中任一项权利要求所述执法分析方法的步骤。
  17. 根据权利要求16所述的计算机可读存储介质，其中，所述执法分析程序被所述处理器执行时实现的步骤S3和步骤S4之间还包括：
    对违法增量变化指数、复议增量变化指数和复议率变化指数根据下式进行加权组合，得到综合指数：
    Figure PCTCN2020105341-appb-100013
    其中，Z为综合指数，W M、W f和W p分别为违法增量变化指数B M、复议增量变化指数B f和复议率变化指数B p的权重,W M+W f+W p＝1，
    其中，在步骤S4中，综合指数与执法健康度正相关。
  18. 根据权利要求17所述的计算机可读存储介质，其中，所述执法分析程序被所述处理器执行时实现的步骤S4包括：
    通过映射函数对综合指数进行数值的映射，获得执法健康度
    E＝f(Z)
    其中，
    Figure PCTCN2020105341-appb-100014
    x＝[-1，1]，f(x)＝[a，b]，[a，b]为映射区间，E为执法健康度，执法健康度与执法健康程度正相关。
  19. 根据权利要求16所述的计算机可读存储介质，其中，所述执法分析程序被所述处理器执行时实现的步骤S1包括：
    将执法人员执法记录仪及执法人员和群众的具有拍摄功能的移动终端拍摄的可能违法的事件的图片上传至图像分析服务器，抽取图像特征；
    将上述图像特征输入卷积神经网络，获得所述图片对应的事件的违法概率；
    将违法概率超过第一设定阈值的事件作为违法事件；
    对不同时刻的违法事件的数量进行统计，进而获得不同时刻的违法量。
  20. 根据权利要求16所述的计算机可读存储介质，其中，所述执法分析程序被所述处理器执行时实现的步骤S3之后还包括：
    通过正态分布对违法增量变化指数、复议增量变化指数、复议率变化指数或/和综合指数进行异常值检测。
PCT/CN2020/105341 2019-08-14 2020-07-29 执法分析方法、装置、电子设备及存储介质 WO2021027569A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910748995.2A CN110580570B (zh) 2019-08-14 2019-08-14 执法分析方法、装置及介质
CN201910748995.2 2019-08-14

Publications (1)

Publication Number Publication Date
WO2021027569A1 true WO2021027569A1 (zh) 2021-02-18

Family

ID=68811180

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/105341 WO2021027569A1 (zh) 2019-08-14 2020-07-29 执法分析方法、装置、电子设备及存储介质

Country Status (2)

Country Link
CN (1) CN110580570B (zh)
WO (1) WO2021027569A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110580570B (zh) * 2019-08-14 2021-01-15 平安国际智慧城市科技股份有限公司 执法分析方法、装置及介质
CN111553816B (zh) * 2020-04-20 2023-11-03 北京北大软件工程股份有限公司 行政复议影响因素分析方法及装置
CN112132455A (zh) * 2020-09-22 2020-12-25 平安国际智慧城市科技股份有限公司 交通执法评价方法、装置、设备及存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106874863A (zh) * 2017-01-24 2017-06-20 南京大学 基于深度卷积神经网络的车辆违停逆行检测方法
CN107992976A (zh) * 2017-12-15 2018-05-04 中国传媒大学 热点话题早期发展趋势预测系统及预测方法
CN108256554A (zh) * 2017-12-20 2018-07-06 深圳市金溢科技股份有限公司 基于深度学习的车辆逆向停车判断方法、服务器及系统
US20190087669A1 (en) * 2017-09-20 2019-03-21 International Business Machines Corporation Management of parking spaces
CN109993056A (zh) * 2019-02-25 2019-07-09 平安科技（深圳）有限公司 一种识别车辆违章行为的方法、服务器及存储介质
CN110580570A (zh) * 2019-08-14 2019-12-17 平安国际智慧城市科技股份有限公司 执法分析方法、装置及介质

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7024360B2 (en) * 2003-03-17 2006-04-04 Rensselaer Polytechnic Institute System for reconstruction of symbols in a sequence
US20130185234A1 (en) * 2012-01-16 2013-07-18 Amrinder Arora System and Method for Using Genetic Algorithm for Optimization of Targeting Systems, Based on Aggregated Scoring Models
CN103164787A (zh) * 2013-02-27 2013-06-19 四川省鼎宸科技发展有限公司 文化执法数字化终端系统和方法
CN103929617A (zh) * 2014-04-14 2014-07-16 李兴海 警察执法现场信息多角度采集传输系统
CN106529902A (zh) * 2016-10-19 2017-03-22 苏州市吴江区公安局 非现场执法人员审核质量检测方法和系统

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106874863A (zh) * 2017-01-24 2017-06-20 南京大学 基于深度卷积神经网络的车辆违停逆行检测方法
US20190087669A1 (en) * 2017-09-20 2019-03-21 International Business Machines Corporation Management of parking spaces
CN107992976A (zh) * 2017-12-15 2018-05-04 中国传媒大学 热点话题早期发展趋势预测系统及预测方法
CN108256554A (zh) * 2017-12-20 2018-07-06 深圳市金溢科技股份有限公司 基于深度学习的车辆逆向停车判断方法、服务器及系统
CN109993056A (zh) * 2019-02-25 2019-07-09 平安科技（深圳）有限公司 一种识别车辆违章行为的方法、服务器及存储介质
CN110580570A (zh) * 2019-08-14 2019-12-17 平安国际智慧城市科技股份有限公司 执法分析方法、装置及介质

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
CHEN SHUSUN: "On the Performance Evaluation Index of Administrative Law Enforcement", JOURNAL OF SOCIAL SCIENCES, 1 January 2014 (2014-01-01), pages 98 - 103, XP055780551, DOI: 10.13644/j.cnki.cn31-1112.2014.03.016 *

Also Published As

Publication number Publication date
CN110580570B (zh) 2021-01-15
CN110580570A (zh) 2019-12-17

Similar Documents

Publication Publication Date Title
WO2021027569A1 (zh) 执法分析方法、装置、电子设备及存储介质
EP3985578A1 (en) Method and system for automatically training machine learning model
WO2021004344A1 (zh) 基于数据分析的风险识别方法及相关设备
CN109993233B (zh) 基于机器学习来预测数据审核目标的方法及系统
WO2021085874A1 (ko) O2o 기반 빅데이터 및 인공지능을 이용한 화물 배차 서비스 제공 방법
CN111652280B (zh) 基于行为的目标物数据分析方法、装置及存储介质
CN108073645B (zh) 一种招聘平台的求职人推荐页面展示方法及装置
WO2022127339A1 (zh) 基于网站注册的用户画像生成方法、装置、设备和介质
CN106095939A (zh) 账户权限的获取方法和装置
CN115545103A (zh) 异常数据识别、标签识别方法和异常数据识别装置
CN114819967A (zh) 数据处理方法、装置、电子设备及计算机可读存储介质
US20220277409A1 (en) Methods and systems for civil administration in smart city based on internet of things (iot)
CN113672643B (zh) 一种业务数字化的分析方法、装置、存储介质及终端
Prathap et al. Crime analysis and forecasting on spatio temporal news feed data—an indian context
Pejic-Bach Profiling intelligent systems applications in fraud detection and prevention: survey of research articles
CN113869427A (zh) 一种场景分析方法、装置、电子设备及存储介质
CN116805039A (zh) 特征筛选方法、装置、计算机设备和数据扰动方法
CN115545349B (zh) 基于属性敏感交互的时序社交媒体流行度预测方法及装置
WO2020202327A1 (ja) 学習システム、学習方法、及びプログラム
JPWO2021048902A1 (ja) 学習モデル適用システム、学習モデル適用方法、及びプログラム
CN116029760A (zh) 消息推送方法、装置、计算机设备和存储介质
JP4938367B2 (ja) セキュリティ診断システム
CN113691525A (zh) 一种流量数据处理方法、装置、设备及存储介质
Sangeetha et al. Crime Rate Prediction and Prevention: Unleashing the Power of Deep Learning
WO2020045107A1 (ja) 評価更新装置、方法、及びプログラム

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20852339

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20852339

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 03/08/2022)

122 Ep: pct application non-entry in european phase

Ref document number: 20852339

Country of ref document: EP

Kind code of ref document: A1