Intelligent supervision method and system based on big data

Abstract

The invention relates to the technical field of intelligent supervision data processing, and discloses an intelligent supervision method and system based on big data, and the system comprises a community intelligent supervision recognition module, a community intelligent supervision processing module, and a community intelligent supervision pushing module. Community violation person registration and non-registration identity type efficient and accurate judgment are carried out based on standard community real-time scene image information in combination with an intelligent search algorithm and community registration person appearance image information stored based on big data; according to the standard community real-time scene image information, in combination with an intelligent recognition algorithm and the identity feature information corresponding to the standard stored appearance image of the community registration personnel, carrying out intelligent search on the identity information of the illegal personnel of the community registration; community intelligent supervision result summary information is scientifically constructed based on registration and non-registration identity type classification of community violation personnel, and the intelligence and quality of community intelligent supervision are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent supervision data processing, and specifically to an intelligent supervision method and system based on big data. Background Art

[0002] Intelligent supervision is widely used in multiple industries such as public management, government supervision, medical supervision, financial supervision, and retail supervision; among them, community intelligent supervision in public management refers to using modern technical means such as the Internet of Things, big data, and artificial intelligence to intelligently monitor and manage public safety, environmental hygiene, and residents' behaviors within the community, thereby improving the efficiency of community governance and the quality of residents' lives. Existing community intelligent supervision operations cannot achieve intelligent supervision of community violation information, nor can they achieve differential and precise processing of community violation information.

[0003] The Chinese invention patent with the publication number CN118350932B discloses a big data model for intelligent financing of small and micro enterprises based on privacy computing and a construction method. By integrating relevant data of government departments and banks, it identifies the characteristics affecting the financing needs of enterprises and the characteristics affecting the loan risks of enterprises; selects the federated learning technology in the privacy computing technology system and relies on the federated learning platform to conduct vertical federated modeling to improve the security of big data for intelligent financing of small and micro enterprises; however, the above technical solutions cannot achieve differential assessment and processing of enterprise financing risks. Summary of the Invention

[0004] (I) Technical Problems to be Solved

[0005] To solve the problems that existing community intelligent supervision operations cannot achieve intelligent supervision of community violation information, nor can they achieve differential and precise processing of community violation information, and to achieve the purposes of accurately identifying community violation behaviors, efficiently judging the identity types of violators, precisely searching for the identity information of violators, scientifically analyzing suggestions for handling community violation behaviors, and differentially and precisely feedbacking the results of community intelligent supervision.

[0006] (II) Technical Solutions

[0007] The present invention is achieved through the following technical solutions: An intelligent supervision method based on big data, the method comprising the following steps:

[0008] S1. Collect real-time community scene image data;

[0009] S2. Perform preprocessing on the image information collected by the community intelligent supervision operation based on the real-time community scene image data, and generate standard real-time community scene image data;

[0010] S3. Perform community violation behavior recognition and processing in community intelligent supervision operations based on the standard community real-time scene image data and community violation behavior image data to generate community violation behavior recognition data. When there is no violation, repeat steps S1, S2, and S3 until a violation occurs;

[0011] S4. When there is a violation, perform community violation personnel identity type judgment and processing in community intelligent supervision operations based on the standard community real-time scene image data and community registered personnel appearance image data to generate community violation personnel identity type judgment data; when the person is not a registered person, directly execute step S6;

[0012] S5. When the person is a registered person, perform community violation personnel identity information search and processing in community intelligent supervision operations based on the standard community real-time scene image data and the corresponding identity feature text data of the community registered personnel appearance image to generate community violation personnel identity feature text data;

[0013] S6. Perform community violation personnel violation behavior handling plan analysis and processing in community intelligent supervision operations based on the community violation behavior recognition data and community violation behavior handling suggestion text data to generate community violation personnel violation behavior handling suggestion text data;

[0014] S7. Construct community intelligent supervision result summary data and perform community intelligent supervision result feedback operations.

[0015] Preferably, the operation steps for collecting community real-time scene image data are as follows:

[0016] S11. Online collect real-time scene image information of personnel activities within the community geographical area through the cloud cameras installed in the community, and generate community real-time scene image data Q.

[0017] Preferably, the operation steps for preprocessing the image information collected in community intelligent supervision operations based on the community real-time scene image data and generating standard community real-time scene image data are as follows:

[0018] S21. Use the BM3D denoising algorithm to perform image denoising preprocessing on the collected community real-time scene image data Q, and generate standard community real-time scene image data Q biaozhun 。

[0019] Preferably, the operation steps for performing community violation behavior recognition and processing in community intelligent supervision operations based on the standard community real-time scene image data and community violation behavior image data to generate community violation behavior recognition data. When there is no violation, repeat steps S1, S2, and S3 until a violation occurs are as follows:

[0020] S31. Establish a community violation behavior image data set where f a represents the community violation behavior image data corresponding to the a-th type of community violation behavior, represents the maximum value of the number of community violation behavior types, and the community violation behavior types include littering, spitting, scribbling, posting advertisements, privately occupying public areas, damaging public facilities, randomly parking non-motor vehicles, and allowing pets to defecate at will; the community violation behavior image data represents the standard scene image information set for community violation behaviors;

[0021] S32. Use the SURF search algorithm to compare the standard community real-time scene image data Q biaozhun with the community violation behavior image data f in the set F of community violation behavior image data a for image feature matching, and generate community violation behavior recognition data F shibie ;

[0022] When Q biaozhun and f a fail to match in image features, indicating that there is no violation behavior in the current community geographical area, then output the community violation behavior recognition data F shibie as non-violation, and at this time, repeat steps S1, S2, and S3 until the community violation behavior recognition data F shibie is a violation;

[0023] When Q biaozhun and f a match successfully in image features, indicating that there is the a-th type of community violation behavior in the current community geographical area, then output the community violation behavior recognition data F shibie as a violation, and at this time, output the text information of the community violation behavior type corresponding to the community violation behavior image data f a ;

[0024] Preferably, when it is a violation, based on the standard community real-time scene image data and the appearance image data of community registered personnel, perform processing for judging the identity type of community violation personnel in community intelligent supervision operations to generate community violation personnel identity type judgment data; when it is a non-registered person, the operation steps of directly executing step S6 are as follows:

[0025] S41. When the community violation behavior recognition data F shibie is a violation, establish a set G=(g1,...,g b ,...,g τ ) of the appearance image data of community registered personnel, where b = 1, 2, 3,..., τ; where g bDenote the appearance image data of the community-registered person corresponding to the b-th community-registered person, and τ denote the maximum value of the number of community-registered persons. The appearance image data of the community-registered person represents the full-body front standard appearance image feature information of the community-registered person;

[0026] S42. Use the ORB search algorithm to process the standard community real-time scene image data Q biaozhun and the appearance image data g of the community-registered person in the set G of the appearance image data of the community-registered person b for image feature matching, and generate judgment data G on the identity type of the community violator based on the image feature matching result panduan ;

[0027] When Q biaozhun and g b fail to match in image features, it means that the violator with violations in the current community geographical area is not a community-registered person. Then output the judgment data G on the identity type of the community violator panduan as a non-registered person, and directly execute step S6 at this time;

[0028] When Q biaozhun and g b match successfully in image features, it means that the violator with violations in the current community geographical area is a community-registered person. Then output the judgment data G on the identity type of the community violator panduan as a registered person.

[0029] Preferably, when it is a registered person, the operation steps for searching and processing the identity information of the community violator in the community intelligent supervision operation based on the corresponding identity feature text data of the standard community real-time scene image data and the appearance image of the community-registered person are as follows:

[0030] S51. Establish a set G'=(g'1,…,g' b ,…,g' τ ) of the corresponding identity feature text data of the appearance image of the community-registered person, where g' b represents the corresponding identity feature text data of the appearance image of the b-th community-registered person. The corresponding identity feature text data of the appearance image of the community-registered person represents the personal identity feature text information of the registered person mapped and set based on the appearance image information of the community-registered person; the personal identity feature text information of the registered person includes the name, residential address, and contact information of the registered person; the contact information includes mobile phone number, WeChat account, and QQ account;

[0031] S52. Use the standard community real-time scene image data Q biaozhunThe identity feature text data g corresponding to the appearance image of the community registered person in the identity feature text data set G' of the community registered person appearance image b Perform image feature matching to search for the standard community real-time scene image data Q biaozhun The identity feature text data g' corresponding to the appearance image of the community registered person that matches b The identity text information of the corresponding registered person, and generate the identity feature text data Y of the community violation person through data identification. The specific operation steps for generating the identity feature text data Y of the community violation person are as follows:

[0032] S521. Initialize parameters. Initialize the spatial dimension in the search space of the identity feature text data set G' of the community registered person appearance image to τ, the manta ray population size for identity information search to N, the maximum number of iterations to T, the standard community real-time scene image data Q biaozhun And all the identity feature text data g' corresponding to the appearance image of the community registered person b The fitness value, and update the position of the manta ray population for identity information search in the search space of the identity feature text data set G' of the community registered person appearance image. The position update formula for the manta ray population for identity information search in the search space of the identity feature text data set G' of the community registered person appearance image is Where i = 1, 2, 3,..., N, Represents the position of the manta ray individual i for identity information search in the search space of the identity feature text data set G' of the community registered person appearance image with a spatial dimension of τ after the (t + 1)-th foraging iteration Represents the position of the manta ray individual i for identity information search in the search space of the identity feature text data set G' of the community registered person appearance image with a spatial dimension of τ after the t-th foraging iteration. α represents the somersault factor, and the value of α is 2. β and χ represent random numbers with values in [0, 1], Represents the position in the search space of the identity feature text data set G' of the community registered person appearance image with a spatial dimension of τ where the manta ray individual for identity information search searches for the identity feature text data g' corresponding to the appearance image of the community registered person that has the maximum fitness value with respect to the standard community real-time scene image data Q biaozhun The identity feature text data g' corresponding to the appearance image of the community registered person with the maximum fitness value b Of the position;

[0033] S522. Initialize the manta ray population for identity information search. The original manta ray individual for identity information search in the manta ray population for identity information search and the identity feature text data g' corresponding to the appearance image of the community registered person in the search space of the identity feature text data set G' of the community registered person appearance image bPerform object matching and search for the standard community real-time scene image data Q biaozhun The identity feature text data g' corresponding to the appearance image of the community registered person with the maximum fitness b Search for manta ray individuals corresponding to the corresponding identity information;

[0034] S523. Determine whether the number of loop iterations meets the maximum iteration number T. If so, go to step S529; otherwise, go to step S524;

[0035] S524. Generate a random number Θ in [0, 1], and perform a comparison of the magnitude of Θ and 0.5. When Θ < 0.5, go to step S525; otherwise, execute steps S521 and S522 to search and generate in the search space of the identity feature text data set G' corresponding to the appearance image of the community registered person that matches the standard community real-time scene image data Q biaozhun The identity feature text data g' corresponding to the appearance image of the community registered person that matches b Search for manta ray individuals corresponding to the corresponding new identity information, and calculate the standard community real-time scene image data Q biaozhun And all the identity feature text data g' corresponding to the appearance image of the community registered person b Of the fitness value;

[0036] S525. Generate a random number Γ in [0, 1], and perform a comparison of the magnitude of Γ and t / T. When Γ < t / T, go to step S526; otherwise, use steps S522 and S523 to search and generate in the search space of the identity feature text data set G' corresponding to the appearance image of the community registered person that matches the standard community real-time scene image data Q biaozhun The identity feature text data g' corresponding to the appearance image of the community registered person that matches b Search for manta ray individuals corresponding to the corresponding new identity information, and calculate the standard community real-time scene image data Q biaozhun And all the identity feature text data g' corresponding to the appearance image of the community registered person b Of the fitness value;

[0037] S526. Search and generate in the search space of the identity feature text data set G' corresponding to the appearance image of the community registered person that matches the standard community real-time scene image data Q according to steps S524 and S525 biaozhun The identity feature text data g' corresponding to the appearance image of the community registered person that matches b Search for manta ray individuals corresponding to the corresponding new identity information, and recalculate the standard community real-time scene image data Q biaozhun And all the identity feature text data g' corresponding to the appearance image of the community registered person bThe fitness value;

[0038] S527. Compare and judge the magnitude of the fitness value of the manta ray individual searched by the new identity information with that of the manta ray individual searched by the original identity information. When the fitness value of the manta ray individual searched by the new identity information is less than that of the manta ray individual searched by the original identity information, replace the manta ray individual searched by the original identity information with the manta ray individual searched by the new identity information; otherwise, retain the manta ray individual searched by the original identity information. Compare and judge the magnitude of the fitness value of the manta ray individual searched by the new identity information with the fitness values of all manta ray individuals searched by identity information in the search space of the set G' of identity feature text data corresponding to the appearance image of the community-registered personnel. When the fitness value of the manta ray individual searched by the new identity information is less than the fitness values of all manta ray individuals searched by identity information in the search space of the set G' of identity feature text data corresponding to the appearance image of the community-registered personnel, then replace the manta ray individual searched by the new identity information with the one corresponding to the standard community real-time scene image data Q in the search space of the set G' of identity feature text data corresponding to the appearance image of the community-registered personnel biaozhun The manta ray individual corresponding to the identity feature text data g' of the appearance image of the community-registered personnel with the maximum fitness b Otherwise, retain the manta ray individual corresponding to the identity feature text data g' of the appearance image of the community-registered personnel with the maximum fitness in the search space of the set G' of identity feature text data corresponding to the appearance image of the community-registered personnel and the standard community real-time scene image data Q biaozhun The manta ray individual corresponding to the identity feature text data g' of the appearance image of the community-registered personnel with the maximum fitness b ;

[0039] S528. Search and generate, according to the steps of S527, the manta ray individual corresponding to the identity feature text data g' of the appearance image of the community-registered personnel that matches the standard community real-time scene image data Q in the search space of the set G' of identity feature text data corresponding to the appearance image of the community-registered personnel biaozhun and recalculate the fitness value of the standard community real-time scene image data Q b with all the identity feature text data g' of the appearance image of the community-registered personnel biaozhun ; b The fitness value;

[0040] S529. When the maximum number of iterations is reached, take the identity text information of the registered personnel corresponding to the identity feature text data g' of the appearance image of the community-registered personnel that best matches the standard community real-time scene image data Q biaozhun and generate, through data identification, the identity feature text data Y of the community violator. b

[0041] Preferably, for the analysis and processing of the community violation handling plan for community violators in the community intelligent supervision operation based on the community violation recognition data and the community violation handling suggestion text data, the operation steps for generating the community violation handling suggestion text data for community violators are as follows:

[0042] S61. Establish a set of community violation handling suggestion text data where k a represents the community violation handling suggestion text data corresponding to the a-th type of community violation behavior, and the community violation handling suggestion text data represents the optimal community violation handling plan information set for community violation behaviors;

[0043] S62. Use the KMP search algorithm to match the keywords of the community violation types between the community violation recognition data F shibie and the community violation handling suggestion text data k in the set of community violation handling suggestion text data K a to search for the community violation handling suggestion text data k shibie that matches the text information of the community violation type corresponding to the community violation recognition data F a , and generate the community violation handling suggestion text data U for community violators through data identification.

[0044] Preferably, the operation steps for constructing the summary data of the community intelligent supervision results and performing the community intelligent supervision result feedback operation are as follows:

[0045] S71. When the community violator identity type judgment data G panduan is a non-registered person, combine and construct the first community intelligent supervision result summary data M1 from the standard community real-time scene image data Q biaozhun , the community violation recognition data F shibie , and the community violation handling suggestion text data U for community violators through data identification, where M1 = (Q biaozhun , F shibie , U);

[0046] When the community violator identity type judgment data G panduan is a registered person, combine and construct the second community intelligent supervision result summary data M2 from the standard community real-time scene image data Q biaozhun , the community violation recognition data F shibie , the community violator identity characteristic text data Y, and the community violation handling suggestion text data U for community violators through data identification, where

[0047] S72. When the community violation person identity type judgment data G panduan is a non-registered person, the community intelligent supervision platform transmits the first community intelligent supervision result summary data M1 to the community public information display terminal through the mobile communication network to perform the community intelligent supervision result feedback operation;

[0048] When the community violation person identity type judgment data G panduan is a registered person, the community intelligent supervision platform transmits the second community intelligent supervision result summary data M2 to the violation person mobile terminal through the mobile communication network based on the community violation person identity feature text data Y to perform the community intelligent supervision result feedback operation.

[0049] An intelligent supervision system based on big data for implementing the described intelligent supervision method based on big data, the system includes a community intelligent supervision recognition module, a community intelligent supervision processing module, and a community intelligent supervision push module;

[0050] The community intelligent supervision recognition module includes a community real-time scene image acquisition unit, a community real-time scene image preprocessing unit, a community violation behavior image storage unit, and a community violation behavior recognition unit;

[0051] The community real-time scene image acquisition unit acquires community real-time scene image data through a cloud lens; the community real-time scene image preprocessing unit preprocesses the image information collected for community intelligent supervision operations based on the community real-time scene image data and generates standard community real-time scene image data; the community violation behavior image storage unit is used to store community violation behavior image data; the community violation behavior recognition unit performs community violation behavior recognition processing in community intelligent supervision operations based on the standard community real-time scene image data and the community violation behavior image data, and generates community violation behavior recognition data;

[0052] The community intelligent supervision processing module includes a community registered person appearance image storage unit, a community violation person identity type judgment unit, a community registered person appearance image corresponding identity feature information storage unit, a community violation person identity feature information search unit, a community violation behavior handling suggestion information storage unit, and a community violation person violation behavior handling suggestion analysis unit;

[0053] The appearance image storage unit of the community registered personnel is used to store the appearance image data of the community registered personnel; the identity type judgment unit of the community violators is used to perform the judgment process of the identity type of the community violators in the community intelligent supervision operation based on the standard community real-time scene image data and the appearance image data of the community registered personnel, and generate the judgment data of the identity type of the community violators; the identity feature information storage unit corresponding to the appearance image of the community registered personnel is used to store the identity feature text data corresponding to the appearance image of the community registered personnel; the identity feature information search unit of the community violators is used to perform the search process of the identity information of the community violators in the community intelligent supervision operation according to the standard community real-time scene image data and the identity feature text data corresponding to the appearance image of the community registered personnel, and generate the identity feature text data of the community violators; the processing suggestion information storage unit of the community violation behavior is used to store the processing suggestion text data of the community violation behavior; the processing suggestion analysis unit of the community violators' violation behavior is used to perform the analysis process of the processing scheme of the community violators' violation behavior in the community intelligent supervision operation according to the community violation behavior recognition data and the processing suggestion text data of the community violation behavior, and generate the processing suggestion text data of the community violators' violation behavior;

[0054] The community intelligent supervision push module includes a community intelligent supervision result information construction unit and a community intelligent supervision result feedback operation execution unit;

[0055] The community intelligent supervision result information construction unit constructs the community intelligent supervision result summary data based on the classification of the identity types of the violators; the community intelligent supervision result feedback operation execution unit executes the community intelligent supervision result feedback operation in categories according to the community intelligent supervision result summary data in combination with the community intelligent supervision platform, the community public information display terminal and the violator mobile terminal.

[0056] (III) Beneficial effects

[0057] The present invention provides an intelligent supervision method and system based on big data. It has the following beneficial effects:

[0058] First, the cloud lens is used to dynamically collect the community real-time scene image information, and the community real-time scene image is preprocessed by noise reduction in combination with the image noise reduction algorithm, so as to improve the accuracy of community intelligent supervision; according to the standard community real-time scene image information, combined with the intelligent search algorithm and the scientifically preset community violation behavior image information, the community violation behavior is accurately identified, realizing the dynamic and efficient supervision of the community violation behavior, and reducing the workload and cost of community intelligent supervision.

[0059] II. By combining the real-time scene image information of the standard community with the intelligent search algorithm and the appearance image information of the community registered personnel stored in the big data, the efficient and reliable judgment of the identity type of the community violators is carried out, realizing the accurate analysis of the registered and unregistered identity types of the community violators, and providing reliable data support for the differential feedback of the community intelligent supervision results; according to the real-time scene image information of the standard community, combining the intelligent recognition algorithm with the corresponding identity feature information of the appearance images of the community registered personnel stored in the standard, the intelligent search of the identity information of the community registered violators is carried out, realizing the efficient and reliable acquisition of the identity feature information of the community registered violators; according to the community violation behavior recognition information, combining the intelligent search algorithm with the text information of the community violation behavior handling suggestions stored in the big data, the accurate matching of the community violation behavior handling plan for the community violators is carried out, realizing the scientific analysis of the community violation behavior handling plan, and improving the intelligence and quality of the community intelligent supervision.

[0060] III. By scientifically constructing the summary information of the community intelligent supervision results based on the classification of the registered and unregistered identity types of the community violators, the summary information of the community intelligent supervision results is generated in a timely manner based on data processing differentiation, improving the effect of the community intelligent supervision; according to the summary information of the community intelligent supervision results, combining the community intelligent supervision platform, the community public information display terminal and the violator mobile terminal to classify and execute the community intelligent supervision result feedback operation, realizing the differential visual feedback of the community intelligent supervision results and improving the applicability of the community intelligent supervision. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a schematic diagram of the modules of a big data-based intelligent supervision system provided by the present invention;

[0062] Figure 2 It is a flowchart of a big data-based intelligent supervision method provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0063] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0064] The embodiments of the big data-based intelligent supervision method and system are as follows:

[0065] Embodiment 1:

[0066] Please refer to Figure 1 - Figure 2 , a big data-based intelligent supervision method, the method includes the following steps:

[0067] S1. Collect real-time scene image data of the community;

[0068] S2. Preprocess the image information collected in the community intelligent supervision operation based on the real-time scene image data of the community, and generate standard real-time scene image data of the community;

[0069] S3. Identify and process community violation behaviors in the community intelligent supervision operation based on the standard real-time scene image data of the community and the community violation behavior image data, and generate community violation behavior identification data. When there is no violation, repeat steps S1, S2, and S3 until there is a violation;

[0070] S4. When there is a violation, judge the identity type of the community violation personnel in the community intelligent supervision operation based on the standard real-time scene image data of the community and the appearance image data of the community registered personnel, and generate community violation personnel identity type judgment data. When it is a non-registered person, directly execute step S6;

[0071] S5. When it is a registered person, search for the identity information of the community violation personnel in the community intelligent supervision operation based on the standard real-time scene image data of the community and the corresponding identity feature text data of the appearance of the community registered personnel, and generate community violation personnel identity feature text data;

[0072] S6. Analyze and process the community violation personnel's violation behavior treatment plan in the community intelligent supervision operation based on the community violation behavior identification data and the community violation behavior treatment suggestion text data, and generate community violation personnel's violation behavior treatment suggestion text data;

[0073] S7. Construct the summary data of the community intelligent supervision results and perform the community intelligent supervision result feedback operation.

[0074] Furthermore, please refer to Figure 1 - Figure 2 , and the operation steps for collecting real-time scene image data of the community are as follows:

[0075] S11. Online collect the real-time scene image information of the activities of personnel within the geographical area of the community through the cloud cameras installed in the community, and generate real-time scene image data Q of the community.

[0076] The operation steps for preprocessing the image information collected in the community intelligent supervision operation based on the real-time scene image data of the community and generating standard real-time scene image data of the community are as follows:

[0077] S21. Use the BM3D denoising algorithm to perform image denoising preprocessing on the collected real-time scene image data Q of the community, and generate standard real-time scene image data Q biaozhun .

[0078] Perform community violation behavior recognition and processing in community intelligent supervision operations based on standard community real-time scene image data and community violation behavior image data, generate community violation behavior recognition data. When it is determined that there is no violation, repeat steps S1, S2, and S3 until a violation occurs. The operation steps are as follows:

[0079] S31. Establish a set of community violation behavior image data where f a represents the community violation behavior image data corresponding to the a-th type of community violation behavior, represents the maximum value of the number of community violation behavior types. Community violation behavior types include littering, spitting, scribbling, posting advertisements, privately occupying public areas, damaging public facilities, randomly parking non-motor vehicles, and allowing pets to defecate at will; Community violation behavior image data represents the standard scene image information set for community violation behaviors;

[0080] S32. Use the SURF search algorithm to match the standard community real-time scene image data Q biaozhun with the community violation behavior image data f in the set of community violation behavior image data F a for image feature matching, and generate community violation behavior recognition data F shibie ;

[0081] When Q biaozhun and f a fail to match in terms of image features, indicating that there is no violation in the current community geographical area, then output the community violation behavior recognition data F shibie as not violated. At this time, repeat steps S1, S2, and S3 until the community violation behavior recognition data F shibie is violated;

[0082] When Q biaozhun and f a match successfully in terms of image features, indicating that there is the a-th type of community violation behavior in the current community geographical area, then output the community violation behavior recognition data F shibie as violated. At this time, output the text information of the community violation behavior type corresponding to the community violation behavior image data f a .

[0083] Through the cooperation of the community real-time scene image acquisition unit and the community real-time scene image preprocessing unit, the community real-time scene image information is dynamically collected by the cloud lens, and the community real-time scene image is preprocessed by combining with the image noise reduction algorithm to improve the accuracy of community intelligent supervision; the community violation behavior recognition unit accurately recognizes community violation behaviors according to the standard community real-time scene image information, combined with the intelligent search algorithm and the scientifically preset community violation behavior image information, realizes the dynamic and efficient supervision of community violation behaviors, and reduces the workload and cost of community intelligent supervision.

[0084] Further, please refer to Figure 1 - Figure 2 , when it is a violation, based on the standard community real-time scene image data and the community registered personnel appearance image data, the community violation personnel identity type judgment process is carried out in the community intelligent supervision operation, and the community violation personnel identity type judgment data is generated; when it is a non-registered personnel, directly execute the operation steps of step S6 as follows:

[0085] S41. When the community violation behavior recognition data F shibie is a violation, establish a community registered personnel appearance image data set G=(g1,…,g b ,…,g τ ), b = 1, 2, 3, …, τ; where g b represents the community registered personnel appearance image data corresponding to the b-th community registered personnel, τ represents the maximum value of the number of community registered personnel, and the community registered personnel appearance image data represents the full-body front standard appearance image feature information of the community registered personnel;

[0086] S42. Use the ORB search algorithm to perform image feature matching between the standard community real-time scene image data Q biaozhun and the community registered personnel appearance image data g b in the community registered personnel appearance image data set G, and generate community violation personnel identity type judgment data G panduan according to the image feature matching result;

[0087] When the image features of Q biaozhun and g b do not match successfully, it means that the violation personnel with violations in the current community geographical area is not a community registered personnel, then output the community violation personnel identity type judgment data G panduan as a non-registered personnel, and directly execute step S6 at this time;

[0088] When the image features of Q biaozhun and g b match successfully, it means that the violation personnel with violations in the current community geographical area is a community registered personnel, then output the community violation personnel identity type judgment data Gpanduan For the registration personnel.

[0089] When being the registration personnel, in the community intelligent supervision operation, for searching and processing the identity information of community violators based on the standard community real-time scene image data and the corresponding identity feature text data of the appearance images of community registration personnel, the operation steps for generating the corresponding identity feature text data of community violators are as follows:

[0090] S51. Establish a set G'=(g'1,…,g' b ,…,g' τ ) of the corresponding identity feature text data of the appearance images of community registration personnel, where g' b represents the corresponding identity feature text data of the appearance image of the b-th community registration personnel. The corresponding identity feature text data of the appearance image of community registration personnel represents the registered personnel's personal identity feature text information mapped and set based on the appearance image information of community registration personnel; the registered personnel's personal identity feature text information includes the name, residential address, and contact information of the registered personnel; the contact information includes mobile phone number, WeChat account, and QQ account;

[0091] S52. Perform image feature matching on the standard community real-time scene image data Q biaozhun and the corresponding identity feature text data g b of the appearance images of community registration personnel in the set G′ of the corresponding identity feature text data of the appearance images of community registration personnel, search for the corresponding identity text information of the registered personnel of the corresponding identity feature text data g′ biaozhun of the appearance images of community registration personnel that matches the standard community real-time scene image data Q b , and generate the corresponding identity feature text data Y of community violators through data identification. The specific operation steps for generating the corresponding identity feature text data Y of community violators are as follows:

[0092] S521. Initialize the parameters, respectively initialize the spatial dimension in the search space of the set G′ of the corresponding identity feature text data of the appearance images of community registration personnel as τ, the manta ray population size for identity information search as N, the maximum number of iterations as T, the fitness values of the standard community real-time scene image data Q biaozhun and all the corresponding identity feature text data g′ b of the appearance images of community registration personnel, and update the positions of the manta ray population for identity information search in the search space of the set G′ of the corresponding identity feature text data of the appearance images of community registration personnel. The position update formula of the manta ray population for identity information search in the search space of the set G′ of the corresponding identity feature text data of the appearance images of community registration personnel is where i = 1, 2, 3,…, N, denotes the position of the identity information search manta ray individual \(i\) in the search space of the set \(G'\) of identity feature text data corresponding to the appearance images of community registered personnel with spatial dimension \(\tau\) after the \((t + 1)\)-th foraging iteration, denotes the position of the identity information search manta ray individual \(i\) in the search space of the set \(G'\) of identity feature text data corresponding to the appearance images of community registered personnel with spatial dimension \(\tau\) after the \(t\)-th foraging iteration, \(\alpha\) represents the loop factor, \(\alpha\) takes the value of 2, and \(\beta\) and \(\chi\) represent random numbers with values in \([0, 1]\), denotes that the identity information search manta ray individual searches in the search space of the set \(G'\) of identity feature text data corresponding to the appearance images of community registered personnel with spatial dimension \(\tau\) to find the one that matches the standard community real-time scene image data \(Q\) biaozhun The identity feature text data \(g'\) corresponding to the appearance image of the community registered personnel with the maximum fitness value b of the position;

[0093] S522. Initialize the identity information search manta ray population, perform object matching between the original identity information search manta ray individuals in the identity information search manta ray population and the identity feature text data \(g'\) corresponding to the appearance images of community registered personnel in the search space of the set \(G'\) of identity feature text data corresponding to the appearance images of community registered personnel, and search for the identity feature text data \(g'\) corresponding to the appearance image of the community registered personnel with the maximum fitness value that matches the standard community real-time scene image data \(Q\) b and search for the identity information search manta ray individual corresponding to the identity feature text data \(g'\) biaozhun of the appearance image of the community registered personnel with the maximum fitness value; b corresponding;

[0094] S523. Determine whether the number of loops meets the maximum number of iterations \(T\). If so, go to step S529; otherwise, go to step S524;

[0095] S524. Generate a random number \(\Theta\) in \([0, 1]\), execute the judgment of comparing the size of \(\Theta\) with 0.5. When \(\Theta\lt0.5\), go to step S525; otherwise, execute steps S521 and S522 to search and generate in the search space of the set \(G'\) of identity feature text data corresponding to the appearance images of community registered personnel the identity feature text data \(g'\) corresponding to the appearance image of the community registered personnel that matches the standard community real-time scene image data \(Q\) biaozhun matched, b and calculate the new identity information search manta ray individual corresponding to the identity feature text data \(g'\) of the standard community real-time scene image data \(Q\) biaozhun and all the identity feature text data \(g'\) b of the appearance images of community registered personnel;

[0096] S525. Generate a random number Γ in [0, 1], and perform a comparison of the numerical values of Γ and t / T. When Γ < t / T, go to step S526; otherwise, use steps S522 and S523 to search in the search space of the set G' of the identity feature text data corresponding to the appearance images of the registered community members to generate the identity feature text data g′ corresponding to the appearance image of the registered community member that matches the standard community real-time scene image data Q biaozhun that matches the standard community real-time scene image data Q b Search for the manta ray individuals corresponding to the new identity information, and calculate the standard community real-time scene image data Q biaozhun and all the identity feature text data g′ corresponding to the appearance images of the registered community members b of the fitness value;

[0097] S526. According to steps S524 and S525, search in the search space of the set G' of the identity feature text data corresponding to the appearance images of the registered community members to generate the identity feature text data g′ corresponding to the appearance image of the registered community member that matches the standard community real-time scene image data Q biaozhun that matches the standard community real-time scene image data Q b Search for the manta ray individuals corresponding to the new identity information, and recalculate the standard community real-time scene image data Q biaozhun and all the identity feature text data g' corresponding to the appearance images of the registered community members b of the fitness value;

[0098] S527. Compare and judge the magnitude of the fitness value of the manta ray individual searched for the new identity information and the fitness value of the manta ray individual searched for the original identity information. When the fitness value of the manta ray individual searched for the new identity information is less than the fitness value of the manta ray individual searched for the original identity information, replace the manta ray individual searched for the original identity information with the manta ray individual searched for the new identity information; otherwise, retain the manta ray individual searched for the original identity information, and compare and judge the magnitude of the fitness value of the manta ray individual searched for the new identity information and the fitness values of all the manta ray individuals searched for the identity information in the search space of the set G′ of the identity feature text data corresponding to the appearance images of the registered community members. When the fitness value of the manta ray individual searched for the new identity information is less than the fitness values of all the manta ray individuals searched for the identity information in the search space of the set G′ of the identity feature text data corresponding to the appearance images of the registered community members, then replace the manta ray individual searched for the identity information corresponding to the identity feature text data g′ with the largest fitness value that matches the standard community real-time scene image data Q in the search space of the set G′ of the identity feature text data corresponding to the appearance images of the registered community members with the manta ray individual searched for the new identity information; otherwise, retain the manta ray individual searched for the identity information corresponding to the identity feature text data g′ with the largest fitness value that matches the standard community real-time scene image data Q in the search space of the set G' of the identity feature text data corresponding to the appearance images of the registered community members biaozhun that matches the standard community real-time scene image data Q b Search for the manta ray individuals corresponding to the new identity information, and calculate the standard community real-time scene image data Q biaozhunThe identity feature text data g′ corresponding to the appearance image of the community registration personnel with the maximum fitness b Search for manta ray individuals corresponding to the corresponding identity information;

[0099] S528. Search and generate the community registration personnel appearance image corresponding identity feature text data g′ that matches the standard community real-time scene image data Q in the search space of the community registration personnel appearance image corresponding identity feature text data set G′ according to step S527 biaozhun The identity feature text data g′ corresponding to the appearance image of the community registration personnel that matches b Search for manta ray individuals corresponding to the corresponding new identity information, and recalculate the standard community real-time scene image data Q biaozhun And all the identity feature text data g' corresponding to the appearance images of the community registration personnel b Of the fitness value;

[0100] S529. When the maximum number of iterations is satisfied, the identity text information of the registration personnel corresponding to the identity feature text data g′ of the appearance image of the community registration personnel that best matches the standard community real-time scene image data Q biaozhun Generate the identity feature text data Y of the community violation personnel through data identification. b The operation steps for analyzing and processing the community violation personnel's violation behavior handling plan according to the community violation behavior recognition data and the community violation behavior handling suggestion text data to generate the community violation personnel's violation behavior handling suggestion text data are as follows:

[0101] S61. Establish a community violation behavior handling suggestion text data set

[0102] Where k Represents the community violation behavior handling suggestion text data corresponding to the a-th type of community violation behavior, and the community violation behavior handling suggestion text data represents the optimal community violation behavior handling plan information set for the community violation behavior; a S62. Use the KMP search algorithm to match the community violation behavior type keywords between the community violation behavior recognition data F

[0103] And the community violation behavior handling suggestion text data k in the community violation behavior handling suggestion text data set K shibie Search for the community violation behavior handling suggestion text data k that matches the community violation behavior type text information corresponding to the community violation behavior recognition data F a And generate the community violation personnel's violation behavior handling suggestion text data U through data identification. shibie a a And generate the community violation personnel's violation behavior handling suggestion text data U through data identification.

[0104] Through the community violation personnel identity type judgment unit, based on the standard community real-time scene image information, combined with the intelligent search algorithm and the community registered personnel appearance image information stored in the big data, the efficient and reliable judgment of the community violation personnel identity type is carried out, realizing the accurate analysis of the registered and unregistered identity types of the community violation personnel, and providing reliable data support for the differential feedback of the community intelligent supervision results; the community violation personnel identity feature information search unit, according to the standard community real-time scene image information, combined with the intelligent recognition algorithm and the corresponding identity feature information of the community registered personnel appearance image stored in the standard, conducts the intelligent search of the identity information of the community registered violation personnel, realizing the efficient and reliable acquisition of the identity feature information of the community registered violation personnel; the community violation personnel violation behavior handling suggestion analysis unit, according to the community violation behavior recognition information, combined with the intelligent search algorithm and the community violation behavior handling suggestion text information stored in the big data, conducts the accurate matching of the community violation personnel violation behavior handling plan, realizing the scientific analysis of the community violation behavior handling plan and improving the intelligence and quality of the community intelligent supervision.

[0105] Further, please refer to Figure 1 - Figure 2 , and the operation steps for constructing the summary data of the community intelligent supervision results and executing the community intelligent supervision result feedback operation are as follows:

[0106] S71. When the community violation personnel identity type judgment data G panduan is an unregistered person, the standard community real-time scene image data Q biaozhun , the community violation behavior recognition data F shibie , and the community violation personnel violation behavior handling suggestion text data U are combined through data identification to construct the first community intelligent supervision result summary data M1, where M1 = (Q biaozhun , F shibie , U);

[0107] When the community violation personnel identity type judgment data G panduan is a registered person, the standard community real-time scene image data Q biaozhun , the community violation behavior recognition data F shibie , the community violation personnel identity feature text data Y, and the community violation personnel violation behavior handling suggestion text data U are combined through data identification to construct the second community intelligent supervision result summary data M2, where M2 = (Q biaozhun , F shibie , Y, U);

[0108] S72. When the community violation personnel identity type judgment data G panduan is an unregistered person, the community intelligent supervision platform transmits the first community intelligent supervision result summary data M1 to the community public information display terminal through the mobile communication network to execute the community intelligent supervision result feedback operation;

[0109] When the identity type judgment data G of community violators panduan is a registered person, the community intelligent supervision platform transmits the second community intelligent supervision result summary data M2 to the violator's mobile terminal through the mobile communication network based on the community violator identity characteristic text data Y to perform the community intelligent supervision result feedback operation.

[0110] Through the community intelligent supervision result information construction unit, the community intelligent supervision result summary information is scientifically constructed based on the classification of the registered and non-registered identity types of community violators, realizing the timely generation of the community intelligent supervision result summary information based on data processing differentiation, and improving the effect of community intelligent supervision; the community intelligent supervision result feedback operation execution unit, according to the community intelligent supervision result summary information, combines the community intelligent supervision platform, the community public information display terminal and the violator's mobile terminal to classify and execute the community intelligent supervision result feedback operation, realizing the differential and visual feedback of the community intelligent supervision result, and improving the applicability of community intelligent supervision.

[0111] Embodiment 2:

[0112] Please refer to Figure 1 - Figure 2 , a big data-based intelligent supervision system for implementing a big data-based intelligent supervision method. The system includes a community intelligent supervision recognition module, a community intelligent supervision processing module, and a community intelligent supervision push module;

[0113] The community intelligent supervision recognition module includes a community real-time scene image acquisition unit, a community real-time scene image preprocessing unit, a community violation behavior image storage unit, and a community violation behavior recognition unit;

[0114] The community real-time scene image acquisition unit acquires community real-time scene image data through a cloud camera; the community real-time scene image preprocessing unit preprocesses the image information collected for the community intelligent supervision operation based on the community real-time scene image data and generates standard community real-time scene image data; the community violation behavior image storage unit is used to store community violation behavior image data; the community violation behavior recognition unit performs community violation behavior recognition processing in the community intelligent supervision operation according to the standard community real-time scene image data and the community violation behavior image data, and generates community violation behavior recognition data;

[0115] The community intelligent supervision processing module includes a community registered person appearance image storage unit, a community violator identity type judgment unit, a community registered person appearance image corresponding identity characteristic information storage unit, a community violator identity characteristic information search unit, a community violation behavior handling suggestion information storage unit, and a community violator violation behavior handling suggestion analysis unit;

[0116] The appearance image storage unit for community registered personnel is used to store the appearance image data of community registered personnel; the identity type judgment unit for community violators conducts the judgment process of the identity type of community violators in the community intelligent supervision operation based on the standard community real-time scene image data and the appearance image data of community registered personnel, and generates the judgment data of the identity type of community violators; the identity feature information storage unit corresponding to the appearance image of community registered personnel is used to store the identity feature text data corresponding to the appearance image of community registered personnel; the identity feature information search unit for community violators conducts the search process of the identity information of community violators in the community intelligent supervision operation according to the standard community real-time scene image data and the identity feature text data corresponding to the appearance image of community registered personnel, and generates the identity feature text data of community violators; the processing suggestion information storage unit for community violation behaviors is used to store the processing suggestion text data for community violation behaviors; the analysis unit for processing suggestions on violation behaviors of community violators conducts the analysis process of the processing plan for violation behaviors of community violators in the community intelligent supervision operation according to the community violation behavior recognition data and the processing suggestion text data for community violation behaviors, and generates the processing suggestion text data for violation behaviors of community violators.

[0117] The community intelligent supervision push module includes a community intelligent supervision result information construction unit and a community intelligent supervision result feedback operation execution unit;

[0118] The community intelligent supervision result information construction unit constructs the summary data of community intelligent supervision results based on the classification of the identity types of violators; the community intelligent supervision result feedback operation execution unit classifies and executes the community intelligent supervision result feedback operation according to the summary data of community intelligent supervision results in combination with the community intelligent supervision platform, the community public information display terminal and the mobile terminal of violators.

[0119] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A big data-based intelligent supervision method, characterized in that, The method includes the following steps: S1. Collect real-time scene image data of the community; S2. Preprocess the image information collected by the community intelligent supervision operation based on the real-time scene image data of the community, and generate standard real-time scene image data of the community; S3. Identify and process community violation behaviors in the community intelligent supervision operation, generate community violation behavior identification data. When there is no violation, repeat steps S1, S2, and S3 until there is a violation; S4. When there is a violation, judge and process the identity type of the community violation personnel in the community intelligent supervision operation, generate community violation personnel identity type judgment data. When the person is an unregistered person, directly execute step S6; S5. When the person is a registered person, search and process the identity information of the community violation personnel in the community intelligent supervision operation, generate community violation personnel identity feature text data; S6. Analyze and process the community violation personnel's violation behavior handling plan in the community intelligent supervision operation, generate community violation personnel's violation behavior handling suggestion text data; S7. Construct the community intelligent supervision result summary data and execute the community intelligent supervision result feedback operation.

2. The intelligent supervision method based on big data according to claim 1, characterized in that: The above S1 includes the following steps: S11. Online collect the real-time scene image information of the activities of personnel in the community geographical area through the cloud cameras installed in the community, and generate the real-time scene image data Q of the community.

3. The intelligent supervision method based on big data according to claim 2, wherein: The above S2 includes the following steps: S21. Use the BM3D noise reduction algorithm to perform image noise reduction preprocessing on the collected Q, and generate standard community real-time scene image data Q biaozhun .

4. The intelligent supervision method based on big data according to claim 3, characterized in that: The above S3 includes the following steps: S31. Establish a set of image data of community violation behaviors where f a represents the image data of community violation behaviors corresponding to the a-th type of community violation behavior, represents the maximum value of the number of types of community violation behaviors; S32. Use the SURF search algorithm to match the Q biaozhun with the f in the F a for image feature matching, and generate community violation behavior recognition data F based on the image feature matching results shibie ; When Q biaozhun and f a do not match successfully in terms of image features, then output the F shibie as non-violating. At this time, repeat steps S1, S2, and S3 until the F shibie is violating; When Q biaozhun matches the image features with f a successfully, the F shibie is determined as a violation, and at this time, the f a corresponding community violation behavior type text information is output.

5. The intelligent supervision method based on big data according to claim 4, wherein: The above S4 includes the following steps: S41. When the said F shibie is in violation, establish a set G=(g1,…,g b ,…,g τ ), b = 1, 2, 3, …, τ of the appearance image data of community registration personnel; where g b represents the appearance image data of the community registration personnel corresponding to the b-th community registration personnel, and τ represents the maximum value of the number of community registration personnel; S42. Use the ORB search algorithm to search for the Q biaozhun and the g in the G b for image feature matching, and generate community violation person identity type judgment data G based on the image feature matching results panduan ; When Q biaozhun and g b do not successfully match any image features, output the G panduan as a non-registered person, and directly execute step S6 at this time; When Q biaozhun successfully matches the image features with g b then output the G panduan as the registered person.

6. The intelligent supervision method based on big data according to claim 5, characterized in that: The above S5 includes the following steps: S51. Establish a set of text data G' = (g'1, …, g' b , …, g' τ ) corresponding to the identity characteristics of the appearance images of community-registered personnel, where g' b represents the text data corresponding to the identity characteristics of the appearance image of the b-th community-registered personnel; S52. Take the Q biaozhun and perform image feature matching with the g' in the G', b search for the g' that matches the Q biaozhun and obtain the identity text information of the registered personnel corresponding to the matched g', b and generate the community violation personnel identity feature text data Y through data identification. The specific operation steps for generating the community violation personnel identity feature text data Y are as follows: S521. Initialize parameters, and initialize the spatial dimension in the search space of the G' to be τ, the population size of the identity information search manta rays to be N, the maximum number of iterations to be T, and the Q biaozhun with all the g' b fitness values, and update the positions of the identity information search manta ray population in the search space of the G'; S522. Initialize the identity information search manta ray population, match the original identity information search manta ray individuals in the identity information search manta ray population with the object in the search space of G', and search for the identity information search manta ray individual corresponding to the g' with the maximum fitness with respect to Q. b of biaozhun the g' with the maximum fitness b ; S523. Judge whether the number of loop times meets the maximum iteration number T. If so, go to step S529; otherwise, go to step S524; S524. Generate a random number Θ in [0, 1], perform a comparison of the magnitude of Θ with 0.

5. When Θ < 0.5, go to step S525; otherwise, execute steps S521 and S522 to search and generate in the search space of G′ the g' that matches the Q biaozhun The corresponding g' that matches b the new identity information to search for manta individuals and calculate the Q biaozhun with all the g′ b fitness values; S525. Generate a random number Γ in [0, 1], perform a comparison of the numerical values of Γ and t / T. When Γ < t / T, go to step S526; otherwise, use steps S522 and S523 to search in the search space of G' to generate the g' biaozhun matching the Q b and calculate the fitness values of the Q biaozhun corresponding new identity information to search for manta individuals, and calculate the Q b with all the g′ S526. Search and generate in the search space of G' according to steps S524 and S525 to obtain g' that matches Q biaozhun and search for manta individuals corresponding to the new identity information of the corresponding g' b and recalculate Q biaozhun and the fitness values of all the g′ b ; S527. Compare and judge the fitness values of the manta ray individuals searched by the new identity information and the fitness values of the manta ray individuals searched by the original identity information. When the fitness value of the manta ray individual searched by the new identity information is less than the fitness value of the manta ray individual searched by the original identity information, replace the manta ray individual searched by the original identity information with the manta ray individual searched by the new identity information; otherwise, retain the manta ray individual searched by the original identity information, and compare and judge the fitness value of the manta ray individual searched by the new identity information with the fitness values of all the manta ray individuals searched by the identity information in the search space of G'. When the fitness value of the manta ray individual searched by the new identity information is less than the fitness values of all the manta ray individuals searched by the identity information in the search space of G', then replace the manta ray individual searched by the identity information corresponding to the g' with the largest fitness value in the search space of G' with the manta ray individual searched by the new identity information; otherwise, retain the manta ray individual searched by the identity information corresponding to the g' with the largest fitness value in the search space of G'. biaozhun The g' with the maximum fitness b corresponding manta ray individual searched by the identity information biaozhun The g' with the maximum fitness b corresponding manta ray individual searched by the identity information S528. Search in the search space of G' according to step S527 to generate g' that matches Q biaozhun and search for manta individuals corresponding to the new identity information of the matched g', and recalculate Q b and calculate the fitness values of all the g' biaozhun ; b ​ S529. When the maximum number of iterations is satisfied, the identity text information of the registration personnel corresponding to the g' that best matches the Q biaozhun is obtained, and through data identification, the identity feature text data Y of the community violator is generated. b ​ 7. The intelligent supervision method based on big data according to claim 6, characterized in that: The above S6 includes the following steps: S61. Establish a collection of text data for handling community violation suggestions where k a represents the text data of the community violation handling suggestion corresponding to the a-th type of community violation S62. Use the KMP search algorithm to match the F shibie with the k in the K a for keyword matching of community violation behavior types, and search for the k shibie that matches the text information of the community violation behavior type corresponding to the F a , and generate the text data U of the processing suggestions for the violation behaviors of community violators through data identification.

8. A big-data-based intelligent supervision method according to claim 7, characterized in that: The above S7 includes the following steps: S71. When the G panduan is a non-registered person, the Q biaozhun , the F shibie , and the U are combined through data identifiers to construct the first community intelligent supervision result summary data M1; When the G panduan is a registration staff, the Q biaozhun , the F shibie , the Y, and the U are combined through data identification to construct the second community intelligent supervision result summary data M2; S72. When the G panduan is a non-registered person, the community intelligent supervision platform transmits the M1 to the community public information display terminal through the mobile communication network to perform the community intelligent supervision result feedback operation; When the G panduan is a registration staff member, the community intelligent supervision platform transmits the M2 to the mobile terminal of the violator through the mobile communication network based on the Y to perform the community intelligent supervision result feedback operation.

9. A big data-based intelligent supervision system for implementing a big data-based intelligent supervision method according to any one of claims 1-8, characterized in that: The system includes a community intelligent supervision recognition module, a community intelligent supervision processing module, and a community intelligent supervision push module.

Images