A community intelligent management method and system based on AI visual recognition

Abstract

The present application discloses a community smart management method and system based on AI visual recognition, which belongs to the field of artificial intelligence. The method includes: capturing facial features through an AI visual recognition module, identifying the facial feature capture results through a preset owner feature database, judging whether there are owner personnel, and if there are owner personnel, judging whether there is personnel feature data of accompanying personnel in the subordinate database, sending an inquiry message to the owner personnel to add associated subordinate data features, and when consent is obtained to add associated subordinate data features, collecting identity information of the accompanying personnel, and associating the identity information collection results of the accompanying personnel with the registration information of the owner personnel. It solves the technical problems of low quality and efficiency of community management, inability to ensure community safety, and low level of intelligent management. It achieves the technical effect of efficient community management based on AI visual recognition, improved management quality, and intelligent community smart management.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence, and in particular to a community intelligent management method and system based on AI visual recognition. Background Art

[0002] With the rapid development of the economy and the steady improvement of social living standards, the quality requirements for smart community management are also increasing. Therefore, studying the combination of community services with new generation information technologies such as the Internet of Things and cloud computing is of great significance for providing people with a safe and smart living environment.

[0003] Currently, communities have access control systems in place, requiring only those with access cards to enter and exit freely. Those without access cards are required to register and confirm with visiting homeowners before they can enter or exit. Furthermore, after entering the community, following the signage to the corresponding building, there are still building controls downstairs, requiring further confirmation with the homeowner. However, the complex access control management process for entering communities and the numerous uncontrollable factors result in inefficient management and unsafe conditions. This leads to technical issues such as low quality and efficiency of community management, an inability to guarantee community safety, and a low level of intelligent management. Summary of the Invention

[0004] The purpose of this application is to provide a community intelligent management method and system based on AI visual recognition to solve the technical problems in the existing technology such as low quality and efficiency of community management, inability to ensure community safety, and low level of intelligent management.

[0005] In view of the above problems, this application provides a community intelligent management method and system based on AI visual recognition.

[0006] In the first aspect, the present application provides a community smart management method based on AI visual recognition, wherein the method is applied to a community smart management system, and the community smart management system is communicatively connected with an AI visual recognition module and a sound collection module, and the method comprises: capturing facial features at a community access control location through the AI ​​visual recognition module, performing feature recognition on the facial feature capture result through a preset owner feature database, and judging whether there is an owner; when the feature recognition result of the facial feature capture result is that there is an owner, performing data comparison on the accompanying personnel of the owner through a subordinate database; judging whether the personnel features of the accompanying personnel exist in the subordinate database; When the personnel feature data does not exist in the subordinate database, feature collection is performed on the accompanying person to obtain an accompanying person feature collection result; an inquiry message for adding associated subordinate data features is sent to the owner person, and feedback sound collection and recognition is performed through the sound collection module; when the feedback sound collection and recognition result matches the voice feature of the owner person and the recognition content is that the associated subordinate data feature is agreed to be added, identity information collection is performed on the accompanying person; the registration information of the owner person is called, and the accompanying person feature collection result is added to the subordinate database, and the identity information collection result of the accompanying person is associated with the registration information of the owner person.

[0007] On the other hand, the present application also provides a community smart management system based on AI visual recognition, wherein the system includes: an AI visual recognition module, the AI ​​visual recognition module is used to capture facial features at the community access control location through the AI ​​visual recognition module, and perform feature recognition on the facial feature capture result through a preset owner feature database to determine whether there is an owner; a data comparison module, the data comparison module is used to perform data comparison of the owner's accompanying personnel through a subordinate database when the feature recognition result of the facial feature capture result is that there is an owner; a feature acquisition module, the feature acquisition module is used to determine whether the subordinate database has the personnel feature data of the accompanying personnel, and when the subordinate database does not have the personnel feature data of the accompanying personnel When the owner obtains the characteristic data of the accompanying person, the characteristics of the accompanying person are collected to obtain the characteristic collection result of the accompanying person; a voice recognition module is used to send an inquiry message for adding the associated subordinate data feature to the owner, and feedback the voice collection and recognition through the voice collection module; an information collection module is used to collect the identity information of the accompanying person when the feedback voice collection and recognition result matches the voice feature of the owner and the recognition content is that the associated subordinate data feature is agreed to be added; an association module is used to call the registration information of the owner, add the characteristic collection result of the accompanying person to the subordinate database, and associate the identity information collection result of the accompanying person with the registration information of the owner. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in this application or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and a person of ordinary skill in the art can obtain other drawings based on the provided drawings without any creative work.

[0009] Figure 1 A flowchart of a community smart management method based on AI visual recognition provided in an embodiment of the present application;

[0010] Figure 2 A schematic diagram of a process for generating temporary traffic data in a community smart management method based on AI visual recognition provided in an embodiment of the present application;

[0011] Figure 3 A flowchart of sending temporary access rights to an electronic device in a community smart management method based on AI visual recognition provided in an embodiment of the present application;

[0012] Figure 4 This is a schematic diagram of the structure of the community smart management system for this application;

[0013] Explanation of the accompanying drawings: AI visual recognition module 11, data comparison module 12, feature acquisition module 13, sound recognition module 14, information acquisition module 15, association module 16. DETAILED DESCRIPTION

[0014] This application provides a community smart management method and system based on AI visual recognition, addressing the technical issues of low quality and efficiency of community management, inability to ensure community safety, and low levels of smart management. It achieves the technical effects of efficient community management based on AI visual recognition, improved management quality, and intelligent community smart management.

[0015] The acquisition, storage, use, and processing of data in this application's technical solution comply with relevant national laws and regulations.

[0016] Below, the technical solutions in this application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited to the example embodiments described herein. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. It should also be noted that, for the convenience of description, only the parts related to this application, rather than all of them, are shown in the accompanying drawings.

[0017] Example 1

[0018] like Figure 1 As shown, the present application provides a community smart management method based on AI visual recognition, wherein the method is applied to a community smart management system, the community smart management system is communicatively connected with an AI visual recognition module and a sound collection module, and the method includes:

[0019] Step S100: The AI ​​visual recognition module is used to capture facial features of people at the community access control location, and feature recognition is performed on the facial feature capture results using a preset owner feature database to determine whether there is an owner.

[0020] Specifically, the AI ​​visual recognition module utilizes machine intelligence to recognize images, replacing manual effort and improving recognition efficiency. The pre-set owner feature database is a pre-set database of facial features representing community owners. This database is used to compare and identify the facial feature capture results obtained by the AI ​​visual recognition module to determine whether an owner is present, thereby achieving the technical effect of improving access control recognition accuracy.

[0021] Step S200: When the feature recognition result of the facial feature capture result indicates that the owner exists, data comparison is performed with the owner's accompanying personnel through the subordinate database;

[0022] Specifically, the subordinate database is a pre-set database containing data on people associated with the owner. Owners can add and delete members from the subordinate database through their mobile devices, thereby managing the information of people associated with them. By comparing the data of the owner's accompanying personnel with the subordinate database, it is possible to determine whether the accompanying personnel are related to the owner, thereby identifying and judging people entering the community, improving the security and efficiency of community management.

[0023] Step S300: determining whether the subordinate database contains the personnel characteristic data of the accompanying person; if the subordinate database does not contain the personnel characteristic data, collecting characteristics of the accompanying person to obtain the accompanying person characteristic collection result;

[0024] Step S400: sending a query message for adding associated subordinate data features to the owner, and performing feedback sound collection and recognition through the sound collection module;

[0025] Step S500: When the feedback sound collection and recognition result matches the voice feature of the owner and the recognition content is that the associated subordinate data feature is agreed to be added, identity information of the accompanying person is collected;

[0026] Specifically, the accompanying person feature collection results include features such as the accompanying person's personal information and their association with the owner. If the accompanying person's personal data features exist in the subordinate database, it indicates that they are associated with the owner and can directly enter the community. If the personal data features do not exist in the subordinate database, a query message is sent to the owner to add the associated subordinate data features. The owner then determines whether to add the accompanying person to the subordinate database, thereby granting them access to the community. Simultaneously, the voice collection module collects and recognizes the owner's feedback, providing an additional layer of security for granting access to the accompanying person. The query message is displayed on the display screen of the AI ​​visual recognition module to query the owner. If the feedback voice collection and recognition results match the owner's voice features, and the recognition content indicates consent to add the associated subordinate data features, it indicates that the permission granting operation was confirmed by the owner, and the collection of the accompanying person's identity information can proceed. This feedback mechanism enhances the security of access control authorization, greatly improves the efficiency of access control recognition, and effectively enhances the service quality of community smart management.

[0027] Step S600: calling the registration information of the owner, adding the feature collection result of the accompanying person to the subordinate database, and associating the identity information collection result of the accompanying person with the registration information of the owner.

[0028] Specifically, the owner's registration information refers to the information recorded by the owner at the management center. This allows the collected characteristics of accompanying personnel to be added to the corresponding subordinate database, ensuring the accuracy of the input information. Furthermore, by associating the collected identity information of accompanying personnel with the owner's registration information, the information of people entering and leaving the community can be recorded, achieving the technical effect of improving community security.

[0029] Further, such as Figure 2 As shown, step S100 in this embodiment of the application further includes:

[0030] Step S110: When the feature recognition result of the facial feature capture result is that the owner does not exist and no matching data of the facial feature capture result is found in the subordinate database, an entry inquiry message is sent to the external user through the AI ​​visual recognition module, wherein the entry inquiry message includes guidance information;

[0031] Step S120: identifying feedback information through the sound collection module, locating the target owner according to the identification result, and generating owner reminder information according to the feedback information;

[0032] Step S130: calling the target owner through the owner reminder information to obtain feedback from the target owner;

[0033] Step S140: When the feedback result is that entry is allowed, temporary pass data is generated and sent to the external user.

[0034] Further, such as Figure 3 As shown, step S140 in this embodiment of the application further includes:

[0035] Step S141: When the feedback result is that the user is allowed to enter, an information reading instruction is sent to the electronic device of the outside user;

[0036] Step S142: performing identity authentication of the external user according to the information reading result and the facial feature capture result to obtain an identity authentication result;

[0037] Step S143: When the identity authentication result is authentication passed, the temporary access data is generated, and the temporary access data includes limited access rights and an offline guidance map;

[0038] Step S144: Send the temporary access authority to the electronic device.

[0039] Furthermore, step S143 of the embodiment of the present application further includes:

[0040] Step S1431: Identifying the associated features of the external user through the AI ​​visual recognition module to obtain external associated features;

[0041] Step S1432: generating path influence parameters according to the external association features;

[0042] Step S1433: determining whether the path impact parameter satisfies a preset level; if the path impact parameter satisfies the preset level, collecting information on available public parking spaces;

[0043] Step S1434: selecting a target parking space based on the vacant public parking space collection result and the location information of the target owner;

[0044] Step S1435: generating a segmented offline guidance map according to the location information of the target parking space and the target owner.

[0045] Specifically, the entry inquiry information is information generated by the AI ​​visual recognition module to inquire about the purpose of a visitor entering the community. Multiple purpose options can be provided, including visiting friends, repairs, and food delivery. The guidance information refers to information guiding the visitor's route after entering the community. The sound collection module then identifies the visitor's response to the entry inquiry information, and based on the identification results, locates the target homeowner corresponding to the visitor. The feedback information is the visitor's response to the entry inquiry information. The target homeowner is the homeowner the visitor intends to visit upon entering the community. The owner reminder message contains the target homeowner information and the reminder content. The target homeowner is then called to obtain their feedback regarding the visitor's request. When entry is granted, temporary access data is generated and sent to the visitor. The temporary access data represents the visitor's access to the community. The temporary access data includes limited-time access control permissions and an offline guidance map. The limited access control permission specifies the number of times the outside user can pass through the community gate. The offline guidance map displays a route map to the target owner on the outside user's electronic device. If normal navigation within the community is not possible due to signal problems, providing offline guidance saves the outside user time.

[0046] Specifically, the electronic device is a mobile device of an external user, and the temporary access data is transmitted by exchanging information between the electronic device and the AI ​​visual recognition module. Optionally, the electronic device can be a mobile phone, a wristband, etc. The method of sending the information reading instruction to the external user's electronic device can optionally achieve information exchange via NFC, Bluetooth, WiFi, or authentication information with WeChat or Alipay.

[0047] Specifically, the information reading result is obtained by reading the information of the outsider's electronic device and the facial feature capture result, and then the identity authentication is performed to determine whether the outsider and the read information are consistent, thereby obtaining an identity authentication result. If the identity authentication result is passed, it can be determined that the outsider's identity is normal. By obtaining the temporary access data generated by the target owner of the target owner, the temporary access permission is obtained through the electronic device.

[0048] Specifically, the AI ​​visual recognition module identifies associated features of outside users. These associated features are obtained to identify influencing parameters for generating an offline guidance map. The externally associated features are characteristic information associated with the outside user's travel information. The path influencing parameters characterize the parameters that influence the outside user's travel within the community. Path information is classified into three levels: car, electric vehicle, and walking. The preset level is pre-set information indicating that the outside user entered by car. When the path influencing parameter meets the preset level, the outside user is driving, and parking space information is further provided to the outside user. Available public parking space information is collected, and a target parking space is selected based on the target owner's location information. The target parking space is selected based on the collected available public parking space information, using the parking space closest to the target owner's location as the target space. Thus, the outside user's movement trajectory within the community is divided into segments of driving and walking, corresponding to the generation of the segmented offline guidance map. This achieves the goal of intelligently distinguishing outside user types and generating corresponding temporary access data, thereby improving access control recognition efficiency and enhancing the intelligent level of community smart management.

[0049] For example, if the feedback from the access inquiry indicates that the incoming user is a food delivery driver and their vehicle is an electric vehicle, the target owner is notified of a food delivery. After the owner sets the number of access control passes, temporary access data for the incoming user is generated. Since this is a food delivery, the access control is generally set to a single pass, and the generated offline guidance map is the optimal route for the electric vehicle.

[0050] For example, if the external user is a friend and the corresponding associated feature is driving, the target owner will be reminded that the friend is visiting. The owner can then provide feedback to allow access and set the number of passes required based on the situation. After collecting available public parking spaces in the community, the system will find the parking space closest to the target owner based on their specific location and provide the external user with an offline map guidance service.

[0051] Furthermore, the community smart management system is also in communication with the early warning module. In the embodiment of the present application, step S100 further includes:

[0052] Step S150: performing feature recognition within the community through the AI ​​visual recognition module to obtain feature recognition results;

[0053] Step S160: determining whether the feature recognition result contains a constraint feature;

[0054] Step S170: When the feature recognition result does not contain the constraint feature, the warning module issues a constraint warning to the associated owner.

[0055] Furthermore, step S170 in the embodiment of the present application further includes:

[0056] Step S171: obtaining basic information of the danger source according to the feature recognition result, wherein the basic information of the danger source includes location information and body shape information;

[0057] Step S172: Identifying user information within the community through the AI ​​visual recognition module to obtain user information, wherein the user information includes location information and basic user information;

[0058] Step S173: generating a relative danger warning level according to the basic information of the danger source and the user information;

[0059] Step S174: issuing a constraint warning to the associated owner through the warning module according to the relative risk warning level.

[0060] Specifically, the community smart management system is also communicatively connected with the early warning module, which is used to provide early warning reminders to community users. The feature recognition within the community through the AI ​​visual recognition module is mainly to identify the sources of danger within the community. The feature recognition result is the result of the source of danger existing in the community. The constraint feature refers to the characteristic information that characterizes the prevention of the source of danger. When the constraint feature does not exist in the feature recognition result, the owner associated with the feature recognition result needs to be subject to a constraint warning through the early warning module. In this way, the goal of associating the source of danger with the responsible owner is achieved. When it is found that no preventive measures have been taken for the source of danger, it can be promptly reported to the associated owner to reduce the probability of danger.

[0061] Specifically, the basic information on dangerous sources refers to information reflecting the basic characteristics of the dangerous source, including location information and size information. The location information indicates the location of the dangerous source within the community, and the size information indicates the size of the dangerous source. The AI ​​visual recognition module identifies user information within the community, obtaining user information that can be used to further analyze the population of the community. User information includes location information and basic user information. Location information refers to the distribution of community users within the community. Basic user information refers to information reflecting the basic characteristics of users, including age and physical condition. This information can be used to determine the user types within the community. The higher the overall user population, the higher the risk of responding to the dangerous source. Furthermore, a relative risk warning level can be generated based on the basic information on dangerous sources and user information. The relative risk warning level represents the risk level of the community and users in responding to the dangerous source. Constraint warnings are issued to associated owners based on different relative risk levels. This allows associated owners to understand the threat posed by their associated dangerous sources to community users, thereby increasing their sense of responsibility and improving community safety.

[0062] Furthermore, step S100 in the embodiment of the present application further includes:

[0063] Step S180: Collect community images using the AI ​​visual recognition module, compare the collected images with historical images, and generate a set of outlier images;

[0064] Step S190: performing abnormal feature recognition on the abnormal point image set, and performing abnormal management within the community based on the abnormal feature recognition results.

[0065] Specifically, by collecting community images and comparing them with historical images of the community, we identify areas that are unusual compared to the past. The historical images refer to the community's conditions over a past period of time, including the operation of public facilities and public spaces. The anomaly point set is a collection of images that characterize differences between the present and past communities. By combining these anomaly point images with anomaly feature recognition, we can identify the anomaly features in the images, and then manage the anomaly areas in a targeted manner. This allows for the timely handling of community anomalies.

[0066] For example, the abnormal feature after the abnormal point image set is identified is that the manhole cover is not covered, which poses a serious safety hazard. After identification, the manhole cover is handled in a timely manner to prevent danger to community users.

[0067] In summary, the community smart management method based on AI visual recognition provided by this application has the following technical effects:

[0068] 1. This application uses an AI visual recognition module to capture facial features. The captured facial features are then identified against a preset owner feature database to determine whether an owner exists. If an owner exists, the application determines whether the accompanying person's profile data exists in the subordinate database. The application then sends a query to the owner requesting the addition of associated subordinate data features. Upon receiving consent to add associated subordinate data features, the application collects the accompanying person's identity information and associates the collected identity information with the owner's registration information. This achieves the technical benefits of efficient community management based on AI visual recognition, improving management quality, and intelligently implementing smart community management.

[0069] 2. This application obtains the relationship between external users and target owners, sets different temporary communication data based on the owners' feedback, and then provides different guidance based on the basic information of external users, thereby achieving the technical effect of improving access control recognition efficiency and improving the intelligence level of community smart management.

[0070] Example 2

[0071] Based on the same inventive concept as the community intelligent management method based on AI visual recognition in the aforementioned embodiment, Figure 4 As shown, the present application also provides a community intelligence management system based on AI visual recognition, wherein the system includes:

[0072] AI visual recognition module 11, the AI ​​visual recognition module 11 is used to capture facial features of people at the community access control location through the AI ​​visual recognition module, perform feature recognition on the facial feature capture results through a preset owner feature database, and determine whether there is an owner;

[0073] A data comparison module 12 is configured to compare data of the owner's entourage with that of the owner through a subordinate database when the feature recognition result of the facial feature capture result indicates that the owner exists;

[0074] A feature collection module 13 is configured to determine whether the subordinate database contains the personnel feature data of the accompanying person. If the subordinate database does not contain the personnel feature data, the module collects features of the accompanying person to obtain a feature collection result of the accompanying person.

[0075] A voice recognition module 14 is used to send an inquiry message to the owner to add associated subordinate data features, and to perform feedback voice collection and recognition through the voice collection module;

[0076] An information collection module 15 is configured to collect identity information of the accompanying person when the feedback voice collection and recognition result is a match with the owner's voice feature and the recognition content is consent to add the associated subordinate data feature;

[0077] The association module 16 is used to call the registration information of the owner, add the feature collection results of the accompanying personnel to the subordinate database, and associate the identity information collection results of the accompanying personnel with the registration information of the owner.

[0078] Furthermore, the AI ​​visual recognition module in the system is also used to:

[0079] When the feature recognition result of the facial feature capture result is that the owner does not exist and the matching data of the facial feature capture result is not found in the subordinate database, an entry inquiry message is sent to the external user through the AI ​​visual recognition module, wherein the entry inquiry message includes guidance information;

[0080] Recognize the feedback information through the sound collection module, locate the target owner according to the recognition result, and generate owner reminder information according to the feedback information;

[0081] Calling the target owner through the owner reminder information to obtain feedback from the target owner;

[0082] When the feedback result is that entry is allowed, temporary pass data is generated and sent to the external user.

[0083] Furthermore, the AI ​​visual recognition module in the system is also used to:

[0084] When the feedback result is that entry is allowed, sending an information reading instruction to the electronic device of the outside user;

[0085] Performing identity authentication of the external user based on the information reading result and the facial feature capture result to obtain an identity authentication result;

[0086] When the identity authentication result is authentication passed, the temporary access data is generated, and the temporary access data includes limited access rights and an offline guidance map;

[0087] The temporary access authority is sent to the electronic device.

[0088] Furthermore, the AI ​​visual recognition module in the system is also used to:

[0089] Identifying the associated features of the external user through the AI ​​visual recognition module to obtain external associated features;

[0090] generating path influence parameters according to the external association features;

[0091] determining whether the path impact parameter satisfies a preset level, and collecting information on available public parking spaces when the path impact parameter satisfies the preset level;

[0092] Selecting a target parking space based on the collected results of available public parking spaces and the location information of the target owner;

[0093] The segmented offline guidance map is generated according to the location information of the target parking space and the target owner.

[0094] Furthermore, the community smart management system is also in communication with an early warning module, and the early warning module in the system is used to:

[0095] Perform feature recognition within the community through the AI ​​visual recognition module to obtain feature recognition results;

[0096] Determining whether the feature recognition result has a constraint feature;

[0097] When the constraint feature does not exist in the feature recognition result, the constraint warning of the associated owner is issued through the warning module.

[0098] Furthermore, the early warning module in the system is used to:

[0099] Obtaining basic information of the danger source according to the feature recognition result, wherein the basic information of the danger source includes location information and body shape information;

[0100] Identify user information within the community through the AI ​​visual recognition module to obtain user information, wherein the user information includes location information and basic user information;

[0101] Generate a relative danger warning level according to the basic information of the danger source and the user information;

[0102] According to the relative risk warning level, the warning module performs a constraint warning for the associated owner.

[0103] Furthermore, the AI ​​visual recognition module in the system is also used to:

[0104] Collect community images through the AI ​​visual recognition module, compare the collected images with historical images, and generate a set of outlier images;

[0105] Abnormal features are identified on the abnormal point image set, and abnormal management within the community is performed based on the abnormal feature identification results.

[0106] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The community smart management method based on AI visual recognition and the specific examples in Example 1 are also applicable to the community smart management system of this embodiment. Through the above detailed description of the community smart management method based on AI visual recognition, those skilled in the art can clearly understand the community smart management system of this embodiment, so for the sake of brevity, it will not be described in detail here. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method description.

[0107] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A community intelligent management method based on AI visual recognition, characterized in that: The method is applied to a community smart management system, which is communicatively connected to an AI visual recognition module and a sound collection module. The method includes: The AI ​​visual recognition module is used to capture facial features at the community access control location, and feature recognition is performed on the facial feature capture results using a preset owner feature database to determine whether there is an owner; When the feature recognition result of the facial feature capture result indicates that the owner exists, performing data comparison with the owner's accompanying personnel through the subordinate database; Determining whether the subordinate database contains the personnel characteristic data of the accompanying person, and when the subordinate database does not contain the personnel characteristic data, collecting characteristics of the accompanying person to obtain a characteristic collection result of the accompanying person; Sending inquiry information for adding associated subordinate data features to the owner personnel, and performing feedback sound collection and recognition through the sound collection module; When the feedback sound collection and recognition result is a match with the owner's voice feature and the recognition content is consent to add the associated subordinate data feature, identity information of the accompanying person is collected; Retrieving the registration information of the owner, adding the feature collection results of the accompanying personnel to the subordinate database, and associating the identity information collection results of the accompanying personnel with the registration information of the owner; The method further comprises: When the feature recognition result of the facial feature capture result is that the owner does not exist and the matching data of the facial feature capture result is not found in the subordinate database, an entry inquiry message is sent to the external user through the AI ​​visual recognition module, wherein the entry inquiry message includes guidance information; Recognize the feedback information through the sound collection module, locate the target owner according to the recognition result, and generate owner reminder information according to the feedback information; Calling the target owner through the owner reminder information to obtain feedback from the target owner; When the feedback result is that entry is allowed, temporary pass data is generated and sent to the external user; The method further comprises: When the feedback result is that entry is allowed, sending an information reading instruction to the electronic device of the outside user; Performing identity authentication of the external user based on the information reading result and the facial feature capture result to obtain an identity authentication result; When the identity authentication result is authentication passed, the temporary access data is generated, and the temporary access data includes limited access rights and an offline guidance map; sending the temporary access authority to the electronic device; The community smart management system is also in communication with the early warning module, and the method further includes: Perform feature recognition within the community through the AI ​​visual recognition module to obtain feature recognition results; Determining whether the feature recognition result has a constraint feature, where the constraint feature is feature information for preventing a danger source; When the constraint feature does not exist in the feature recognition result, a constraint warning is issued to the associated owner through the warning module; The method further comprises: Obtaining basic information of the danger source according to the feature recognition result, wherein the basic information of the danger source includes location information and body shape information; Identify user information within the community through the AI ​​visual recognition module to obtain user information, wherein the user information includes location information and user basic information, and the user basic information is information reflecting the basic situation of the user, including age and physical condition; Generate a relative danger warning level according to the basic information of the danger source and the user information; According to the relative risk warning level, the warning module performs a constraint warning for the associated owner.

2. The method according to claim 1, wherein The method further comprises: Identifying the associated features of the external user through the AI ​​visual recognition module to obtain external associated features; generating path influence parameters according to the external association features; determining whether the path impact parameter satisfies a preset level, and collecting information on available public parking spaces when the path impact parameter satisfies the preset level; Selecting a target parking space based on the collected results of available public parking spaces and the location information of the target owner; The segmented offline guidance map is generated according to the location information of the target parking space and the target owner.

3. The method according to claim 1, wherein The method further comprises: Collect community images through the AI ​​visual recognition module, compare the collected images with historical images, and generate a set of outlier images; Abnormal features are identified on the abnormal point image set, and abnormal management within the community is performed based on the abnormal feature identification results.

4. A community intelligence management system based on AI visual recognition, characterized by: The system is used to perform the method according to any one of claims 1 to 3, and the system includes: AI visual recognition module, which is used to capture facial features of people at community access control locations through the AI ​​visual recognition module, perform feature recognition on the facial feature capture results through a preset owner feature database, and determine whether there is an owner; A data comparison module, configured to compare data of the owner's entourage through a subordinate database when a feature recognition result of the facial feature capture result indicates that the owner exists; a feature collection module, the feature collection module being used to determine whether the subordinate database contains the personnel feature data of the accompanying person, and when the subordinate database does not contain the personnel feature data, collecting features of the accompanying person to obtain a feature collection result of the accompanying person; A voice recognition module, the voice recognition module is used to send an inquiry message with associated subordinate data features to the owner, and perform feedback voice collection and recognition through the voice collection module; An information collection module, which is used to collect the identity information of the accompanying person when the feedback sound collection and recognition result matches the voice feature of the owner and the recognition content is to agree to add the associated subordinate data feature; an association module, which is used to call the registration information of the owner, add the accompanying person feature collection result to the subordinate database, and associate the accompanying person's identity information collection result with the owner's registration information.

Images