Real-time property safety monitoring system for smart city

The real-time property safety monitoring system addresses inefficiencies in traditional smart city security by automating face recognition and risk assessment, enhancing community safety through intelligent supervision and risk management.

CN119091383APending Publication Date: 2024-12-06CHONGQING TONGYI PROPERTY MANAGEMENT CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411279579.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-12
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

Traditional property safety monitoring in smart cities is ineffective when homeowners are away or mobile terminals malfunction, lacks comprehensive community monitoring, and fails to integrate with police for enhanced security.

Method used

A real-time property safety monitoring system comprising a monitoring center connected to data acquisition, processing, face recognition, and security management modules, which collects and processes user information, performs face recognition, and generates supervision instructions for risk or security identification.

Benefits of technology

The system provides automated face identification, posture estimation, and risk assessment, improving monitoring efficiency and security management by generating timely warnings and intelligent risk responses.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119091383A_ABST
    Figure CN119091383A_ABST
Patent Text Reader

Abstract

The invention discloses a real-time property safety monitoring system for a smart city, and relates to the field of safety monitoring, the real-time property safety monitoring system comprises a monitoring center, and the monitoring center is in communication connection with a data acquisition module, a data processing module, a face recognition module and a safety management module; the data acquisition module is used for dividing a target property area to obtain a plurality of property areas, and performing data acquisition on the obtained property areas to obtain corresponding user basic information and video information; the data processing module is used for processing the collected basic information of the user and constructing a corresponding face database according to a processing result; the face recognition module is used for performing face recognition on a target property area based on the constructed face database and generating a corresponding supervision instruction according to a recognition result, and the safety management module is used for taking corresponding supervision measures according to the obtained supervision instruction; according to the invention, the efficiency and accuracy of property management are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of security monitoring, and in particular to a real-time property security monitoring system for a smart city. Background Art

[0002] Smart city refers to a model that uses various intelligent technologies and methods to integrate various existing service resources in the city to provide urban residents with a variety of convenient services such as government affairs, business, entertainment, education, medical care, and mutual assistance in life. The setting of smart city provides residents in the community with a faster and more comfortable intelligent living environment. In the process of building smart city, security monitoring system also plays a very important role. Compared with the existing technology, traditional property safety monitoring is mostly conducted by security personnel for on-site inspections, or mainly focuses on home security. If the homeowner is away or there is a malfunction on the mobile terminal, the security system will be ineffective, and the safety hazards are quite large. In addition, the overall monitoring method is relatively simple and cannot comprehensively monitor community safety. The overall security effect is not good, and it is not convenient to cooperate with the police, and good security effects cannot be achieved. These are the problems we need to solve. For this reason, we provide a real-time property safety monitoring system for smart cities. Summary of the invention

[0003] The purpose of the present invention is to provide a real-time property safety monitoring system for a smart city.

[0004] The object of the present invention can be achieved by the following technical solutions: A real-time property safety monitoring system for a smart city, comprising a monitoring center, wherein the monitoring center is communicatively connected to a data acquisition module, a data processing module, a face recognition module, and a safety management module; The data collection module is used to divide the target property area, obtain a number of property areas, and collect data on the obtained property areas to obtain corresponding user basic information and video information; The data processing module is used to process the collected basic information of the user and build a corresponding face database according to the processing results; The face recognition module is used to perform face recognition on the target property area based on the constructed face database, and generate corresponding supervision instructions according to the recognition results, and the supervision instructions include risk instructions and safety recognition instructions; The security management module is used to take corresponding supervision measures according to the obtained supervision instructions.

[0005] Furthermore, the data collection module divides the target property area to obtain a plurality of property areas, and collects data from the obtained property areas. The process of obtaining corresponding user basic information and video information includes: Obtain a bird's-eye view corresponding to the corresponding smart city, and obtain the property area to be monitored based on it, divide the property area to be monitored into unequal parts, and obtain several property communities; The data collection module is provided with a number of collection terminals and monitoring terminals. The collection terminals are used for permanent users in the corresponding property community to input information, obtain corresponding user basic information, and upload it to the monitoring center for review. After the review is passed, the monitoring center stores the user basic information that has passed the review; The monitoring terminal is used to collect video of the corresponding property community and obtain corresponding video information.

[0006] Furthermore, the process of the data processing module for processing the collected basic user information includes: Read the collected basic information of users and generate corresponding community codes based on it; Obtain a facial image corresponding to a corresponding user, and obtain all pixel points in the corresponding facial image; Constructing a two-dimensional rectangular coordinate system, and mapping the corresponding face image into the constructed two-dimensional rectangular coordinate system, obtaining pixel point coordinates and a corresponding pixel point set in the corresponding face image; Reduce the pixel point coordinates in the obtained pixel point set to obtain a corresponding reduced point set; After the reduction is completed, convolution is performed on the coordinates of the corresponding reduced pixel point with any pixel point as the center; after the convolution is completed, the corresponding convolution image is obtained, the pixel values ​​corresponding to the pixel points in the corresponding convolution image are obtained and normalized to obtain the corresponding probability image.

[0007] Furthermore, the process of constructing a corresponding face database based on the processing results includes: Set a probability threshold, and compare the pixel value corresponding to the pixel point in the corresponding probability image with the probability threshold; If the pixel value is lower than the probability threshold, no other operation is performed; if the pixel value is not lower than the probability threshold, the corresponding pixel point in the associated face image is marked as a key recognition point; Obtaining pixel coordinates corresponding to corresponding key recognition points in the face image, and marking them as recognition coordinates; Acquire feature data between corresponding key recognition points based on the obtained recognition coordinates, combine the feature data corresponding to the corresponding face images, and obtain a corresponding face feature vector; All collected face images are processed using the same method to obtain corresponding face feature vectors; at the same time, the community code corresponding to the corresponding user is obtained, and the community code is associated with the obtained feature data and face feature vector; after the association is completed, a face database is constructed based on the obtained face feature vector.

[0008] Furthermore, the process of the face recognition module processing the collected video information to obtain the corresponding image to be recognized includes: Read the video information uploaded in real time, divide the obtained video information frame by frame, and obtain several video frame images; Based on face recognition technology, face recognition is performed frame by frame on the obtained video frame images according to the time sequence. If no corresponding face information is recognized in the corresponding video frame, no other operation is performed; If facial information is recognized, the recognized person is marked as a user to be recognized, and other video frame images adjacent to the corresponding video frame image are read with the corresponding video frame image as the center; All video frame images are processed using the same method to obtain key identification points corresponding to the corresponding video frame images, and recorded as key feature points; Based on the obtained key feature points, the face information in the corresponding video frame image is estimated to obtain the corresponding face pose; The facial posture obtained in the corresponding video frame image is compared according to the time before and after, and whether the facial posture in the corresponding video image has changed is determined based on the comparison result; if changed, the corresponding video frame image is cropped, enlarged and corrected based on the obtained key feature points to obtain the corresponding image to be identified.

[0009] Furthermore, the process of performing face recognition on the corresponding image to be recognized in combination with the obtained face database includes: The same method is used to obtain feature data and corresponding facial feature vectors of the image to be identified; Performing face retrieval in a corresponding face database based on the obtained face feature vector and obtaining a corresponding similarity coefficient; A similarity threshold is set, and the obtained similarity coefficient is compared with the similarity threshold; and corresponding supervision instructions are generated according to the comparison result, and the supervision instructions include risk instructions and safety identification instructions.

[0010] Furthermore, the process of taking corresponding supervisory measures based on the risk instructions obtained includes: If the supervision instruction is a risk instruction, the real-time position of the corresponding person to be identified in the corresponding property community is obtained based on the obtained video frame image, and it is fed back to the monitoring center. The monitoring center takes the real-time position of the corresponding person to be identified as the center, notifies the nearest security personnel to patrol the corresponding person to be identified, and chooses whether to expel the corresponding person to be identified or call the police based on the patrol results.

[0011] Further, the process of taking corresponding measures according to the obtained security identification instruction includes: If the corresponding supervision instruction is a security identification instruction, obtain the facial feature vector of the corresponding suspicious person; and based on it, search the video information collected by the corresponding monitoring terminal, and according to the search result, collect the corresponding information to the monitoring terminal related to the corresponding suspicious person, and mark it as an identification terminal; Obtain the real-time location and monitoring range of all identification terminals, and based on them, obtain the visit area ratio, security level and stay time corresponding to the corresponding suspicious persons; Conduct risk assessment on the corresponding suspicious persons based on the obtained visit area ratio, security level and stay time, and obtain corresponding risk scores; Set a risk threshold range and compare the obtained risk score with the risk threshold; If the risk score is lower than the minimum value of the risk threshold range, no other operation is performed; if the risk score is within the risk threshold range, the real-time location of the corresponding suspicious person is obtained in real time based on the monitoring terminal, and security patrols in the corresponding area are strengthened; If the risk score is higher than the maximum value of the risk threshold range, a risk warning will be fed back to the monitoring center, and the corresponding monitoring center will assign security personnel to verify the identity and patrol the corresponding suspicious persons.

[0012] Compared with the prior art, the present invention has the following beneficial effects: 1. The real-time video monitoring system can automatically identify the face information in the video, and perform posture estimation and feature extraction; by analyzing the video frame by frame, the system can accurately identify the identity of the person and give timely warnings for abnormal situations; it not only improves the monitoring efficiency, but also reduces manual intervention, providing strong support for property safety management; 2. Through the linkage between the visitor database and the monitoring terminal, intelligent identification and risk assessment of unregistered visitors are realized; risk instructions or security identification instructions are automatically generated based on the assessment results, and security personnel are notified to take corresponding measures; the safety management level of the property community is effectively improved, and a strong guarantee is provided for preventing and responding to potential security risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0014] like Figure 1 As shown, a real-time property safety monitoring system for a smart city includes a monitoring center, wherein the monitoring center is communicatively connected to a data acquisition module, a data processing module, a face recognition module, and a safety management module; The data collection module is used to divide the target property area, obtain a number of property areas, and collect data on the obtained property areas to obtain corresponding user basic information and video information; The data processing module is used to process the collected basic information of the user and build a corresponding face database according to the processing results; The face recognition module is used to perform face recognition on the target property area based on the constructed face database, and generate corresponding supervision instructions according to the recognition results, and the supervision instructions include risk instructions and safety recognition instructions; The security management module is used to take corresponding supervision measures according to the obtained supervision instructions; It should be further explained that, in the specific implementation process, the data collection module divides the target property area to obtain a number of property communities, and collects data on the obtained property areas. The process of obtaining corresponding property information includes: Obtain a bird's-eye view corresponding to the corresponding smart city, and obtain the property area to be monitored based on the bird's-eye view, divide the property area to be monitored unequally, and obtain a number of property communities, wherein the property community may be an industrial park or a residential area under the corresponding smart city; The data collection module is provided with a number of collection terminals and monitoring terminals. The collection terminals are used for permanent users in the corresponding property community to input information, obtain corresponding user basic information, and upload it to the monitoring center for review. After the review is passed, the monitoring center stores the user basic information that has passed the review; The user's basic information includes but is not limited to ID number, mobile phone number, name, courtyard, building, unit, floor, household and facial image; The monitoring terminal is deployed in the corresponding property community according to actual needs, and is used to collect video of the corresponding property community and obtain corresponding video information; The collected basic user information and video information are uploaded to the monitoring center and stored.

[0015] It should be further explained that, in the specific implementation process, the data processing module processes the collected basic information of the user, and the process of constructing a corresponding face database according to the processing results includes: Read the collected basic information of the user and generate a corresponding community code based on it. The community code is used to uniquely authenticate the corresponding user in the property community; Obtain a facial image corresponding to the corresponding user, obtain all pixels in the corresponding facial image, and number them, denoted as i, where i=1, 2, ..., n, n>0 and n is an integer, and n is the total number of pixels in the corresponding facial image; Construct a two-dimensional rectangular coordinate system, and map the corresponding face image into the constructed two-dimensional rectangular coordinate system, then the pixel coordinates in the corresponding face image are Si (x, y); then the pixel set in the corresponding face image can be expressed as {Si (x, y) | 0 ≤ x, y < 256, x, y are integers}; Reduce the pixel point coordinates in the corresponding pixel point set to obtain a corresponding reduced point set, where the corresponding reduced point set is represented by {Pi(X, Y) | 0≤X, Y<256 / a, X, Y are integers}; where a represents the multiple of the corresponding pixel point coordinate reduction; After the reduction is completed, the coordinates of the corresponding reduced pixel points are Gaussian convolved with the coordinates Pi (X, Y) as the center. The corresponding mathematical formula is as follows: ; Where Gi (X, Y) represents the pixel coordinates after Gaussian convolution of pixel i; After the convolution is completed, the corresponding convolution image is obtained, the pixel values ​​corresponding to the pixel points in the corresponding convolution image are obtained and normalized to obtain the corresponding probability image; Correlating the obtained probability image with the pixel points in the corresponding face image according to the pixel point coordinates; Set a probability threshold, and compare the pixel value corresponding to the pixel point in the corresponding probability image with the probability threshold; If the pixel value is lower than the probability threshold, no other operation is performed; if the pixel value is not lower than the probability threshold, the corresponding pixel point in the associated face image is marked as a key recognition point; Obtaining pixel coordinates corresponding to corresponding key recognition points in the face image, and marking them as recognition coordinates; Based on the obtained identification coordinates, feature data between corresponding key identification points are obtained, and the feature data include but are not limited to the relative position, Euclidean distance, curvature and angle between the key identification points; it should be further explained that the process of obtaining the feature data is a prior art and will not be elaborated in detail in the present invention; Combining the feature data corresponding to the corresponding face images to obtain a corresponding face feature vector, wherein the combination of the feature data refers to encoding the relative position and angle information between each key point into a vector form according to a preset rule; All collected face images are processed using the same method to obtain corresponding face feature vectors; at the same time, the community code corresponding to the corresponding user is obtained, and the community code is associated with the obtained feature data and face feature vector; wherein, the community code can be used to quickly retrieve and obtain the basic user information corresponding to the corresponding user in the monitoring center; after the association is completed, a face database is constructed based on the obtained face feature vector.

[0016] It should be further explained that, in the specific implementation process, the face recognition module performs face recognition on the target property area based on the constructed face database, and generates corresponding instructions according to the recognition result, including: Read the video information uploaded in real time, divide the obtained video information frame by frame, obtain a number of video frame images, and perform face recognition on the obtained video frame images frame by frame according to the time sequence based on face recognition technology. If the corresponding face information is not recognized in the corresponding video frame, no other operations are performed; If facial information is recognized, the recognized person is marked as a user to be recognized, and at the same time, with the corresponding video frame image as the center, N other video frame images adjacent to the corresponding video frame image are read; where N is an integer and N is determined according to actual needs; All video frame images are processed using the same method to obtain key identification points corresponding to the corresponding video frame images, and recorded as key feature points; Then, based on the obtained key feature points, the face information in the corresponding video frame image is estimated to obtain the corresponding face pose, which includes but is not limited to the face pitch angle, yaw angle and roll angle; The facial postures obtained in the corresponding video frame images are compared according to the time before and after, and whether the facial postures in the corresponding video images have changed is determined according to the comparison results. If there is no change, it indicates that the corresponding person to be identified is not a real person, and no other operations are performed; if there is a change, it indicates that the corresponding person to be identified is a real person, and the corresponding video frame images are cropped, enlarged and corrected based on the obtained key feature points to obtain the corresponding image to be identified; The obtained image to be identified is processed by the same method to obtain corresponding feature data and face feature vector; The obtained images to be identified are numbered and recorded as j, where j=1, 2, ..., m, m>0 and m is an integer, and m is the total number of the obtained images to be identified; then the facial feature vector corresponding to the corresponding image to be identified is recorded as Ti; Based on the obtained face feature vector Ti, face retrieval is performed in the corresponding face database, and the corresponding similarity coefficient is obtained, and the obtained similarity coefficient is recorded as XS; in,

[0017] Wherein, G represents the face feature vector of the face image corresponding to a resident user in the face database; k represents the dimension of the corresponding face feature vector, and the present invention defaults to k=128; Set a similarity threshold, compare the obtained similarity coefficient with the similarity threshold; and generate corresponding regulatory instructions based on the comparison results; If the similarity coefficient is not lower than the similarity threshold, it indicates that the corresponding person to be identified is a resident user, and no other operations are performed; If the similarity coefficient is lower than the similarity threshold, indicating that the corresponding person to be identified is a non-resident user, the same method is used to compare the facial feature vector Ti corresponding to the corresponding person to be identified with the facial feature vector stored in the corresponding visitor database; and based on the comparison result, it is determined whether it exists in the visitor database. If it does not exist in the visitor database, a corresponding risk instruction is generated; if it exists in the visitor database, a corresponding security identification instruction is generated, and the corresponding person to be identified is marked as a suspicious person; wherein, the visitor database is constructed based on the basic user information registered by the visitors of the corresponding property community.

[0018] It should be further explained that, in the specific implementation process, the process in which the security management module takes corresponding supervision measures according to the obtained supervision instructions includes: Read the obtained supervision instruction. If the supervision instruction is a risk instruction, obtain the real-time position of the corresponding person to be identified in the corresponding property community based on the obtained video frame image, and feed it back to the monitoring center. Then, the monitoring center notifies the nearest security personnel to patrol the corresponding person to be identified based on the real-time position of the corresponding person to be identified, and can choose whether to expel the corresponding person to be identified or call the police based on the patrol results; If the corresponding supervision instruction is a security identification instruction, obtain the facial feature vector of the corresponding suspicious person; and based on it, search the video information collected by the corresponding monitoring terminal, and according to the search result, collect the corresponding information to the monitoring terminal related to the corresponding suspicious person, and mark it as an identification terminal; Obtain the real-time location and monitoring range corresponding to all identification terminals, and based on them, obtain the access area of ​​the corresponding suspected person and the corresponding access area area; compare the corresponding access area area with the total area of ​​the corresponding property community to obtain the corresponding access area ratio; it should be further explained that different models of monitoring terminals correspond to different monitoring ranges; Based on the real-time location corresponding to the corresponding identification terminal, the monitoring range corresponding to the corresponding identification terminal is divided into security levels to obtain the corresponding security level. For example, the closer to the fence or substation in the property community, the higher the security level corresponding to the corresponding identification terminal; At the same time, the corresponding stay time of the corresponding suspected person at the corresponding identification terminal is obtained according to the search results; Then, based on the obtained visit area ratio, security level and stay time, risk assessment is conducted on the corresponding suspicious persons to obtain corresponding risk scores; In one embodiment of the present invention, for example, the higher the percentage of the area visited by the corresponding suspicious person or the longer the person stays in the area with a high security level, the higher the corresponding risk score; Set the risk threshold range and compare the obtained risk score with the risk threshold; If the risk score is lower than the minimum value of the risk threshold range, it means that the corresponding suspicious person does not pose a risk, and no other operations are performed; If the risk score is within the risk threshold range, the real-time location of the corresponding suspicious person is obtained in real time based on the monitoring terminal, and security patrols in the corresponding area are strengthened; If the risk score is higher than the maximum value of the risk threshold range, a risk warning will be fed back to the monitoring center, and the corresponding monitoring center will assign security personnel to verify the identity and patrol the corresponding suspicious persons.

[0019] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A real-time property safety monitoring system for smart cities, characterized in that: It includes a monitoring center, which is communicatively connected to a data acquisition module, a data processing module, a face recognition module and a security management module; The data collection module is used to divide the target property area, obtain a number of property areas, and collect data on the obtained property areas to obtain corresponding user basic information and video information; The data processing module is used to process the collected basic information of the user and build a corresponding face database according to the processing results; The face recognition module is used to perform face recognition on the target property area based on the constructed face database, and generate corresponding supervision instructions according to the recognition results, and the supervision instructions include risk instructions and safety recognition instructions; The security management module is used to take corresponding supervision measures according to the obtained supervision instructions.

2. A real-time property safety monitoring system for smart cities according to claim 1, characterized in that: The data collection module divides the target property area to obtain a number of property areas, and collects data from the obtained property areas to obtain the corresponding user basic information and video information. The process includes: Obtain a bird's-eye view corresponding to the corresponding smart city, and obtain the property area to be monitored based on it, divide the property area to be monitored into unequal parts, and obtain several property communities; The data collection module is provided with a number of collection terminals and monitoring terminals. The collection terminals are used for permanent users in the corresponding property community to input information, obtain corresponding user basic information, and upload it to the monitoring center for review. After the review is passed, the monitoring center stores the user basic information that has passed the review; The monitoring terminal is used to collect video of the corresponding property community and obtain corresponding video information.

3. A real-time property safety monitoring system for smart cities according to claim 2, characterized in that: The process of the data processing module for processing the collected basic user information includes: Read the collected basic information of users and generate corresponding community codes based on it; Obtain a facial image corresponding to a corresponding user, and obtain all pixel points in the corresponding facial image; Constructing a two-dimensional rectangular coordinate system, and mapping the corresponding face image into the constructed two-dimensional rectangular coordinate system, obtaining pixel point coordinates and a corresponding pixel point set in the corresponding face image; Reduce the pixel point coordinates in the obtained pixel point set to obtain a corresponding reduced point set; After the reduction is completed, convolution is performed on the coordinates of the corresponding reduced pixel point with any pixel point as the center; after the convolution is completed, the corresponding convolution image is obtained, the pixel values ​​corresponding to the pixel points in the corresponding convolution image are obtained and normalized to obtain the corresponding probability image.

4. A real-time property safety monitoring system for smart cities according to claim 3, characterized in that: The process of building a corresponding face database based on the processing results includes: Set a probability threshold, and compare the pixel value corresponding to the pixel point in the corresponding probability image with the probability threshold; If the pixel value is lower than the probability threshold, no other operation is performed; if the pixel value is not lower than the probability threshold, the corresponding pixel point in the associated face image is marked as a key recognition point; Obtaining pixel coordinates corresponding to corresponding key recognition points in the face image, and marking them as recognition coordinates; Acquire feature data between corresponding key recognition points based on the obtained recognition coordinates, combine the feature data corresponding to the corresponding face images, and obtain a corresponding face feature vector; All collected face images are processed using the same method to obtain corresponding face feature vectors; at the same time, the community code corresponding to the corresponding user is obtained, and the community code is associated with the obtained feature data and face feature vector; after the association is completed, a face database is constructed based on the obtained face feature vector.

5. A real-time property safety monitoring system for smart cities according to claim 3, characterized in that: The process of the face recognition module processing the collected video information to obtain the corresponding image to be recognized includes: Read the video information uploaded in real time, divide the obtained video information frame by frame, and obtain several video frame images; Based on face recognition technology, face recognition is performed frame by frame on the obtained video frame images according to the time sequence. If no corresponding face information is recognized in the corresponding video frame, no other operation is performed; If facial information is recognized, the recognized person is marked as a user to be recognized, and other video frame images adjacent to the corresponding video frame image are read with the corresponding video frame image as the center; All video frame images are processed using the same method to obtain key identification points corresponding to the corresponding video frame images, and recorded as key feature points; Based on the obtained key feature points, the face information in the corresponding video frame image is estimated to obtain the corresponding face pose; The facial posture obtained in the corresponding video frame image is compared according to the time before and after, and whether the facial posture in the corresponding video image has changed is determined based on the comparison result; if changed, the corresponding video frame image is cropped, enlarged and corrected based on the obtained key feature points to obtain the corresponding image to be identified.

6. A real-time property safety monitoring system for smart cities according to claim 5, characterized in that: The process of performing face recognition on the corresponding image to be recognized in combination with the obtained face database includes: The same method is used to obtain feature data and corresponding facial feature vectors of the image to be identified; Performing face retrieval in a corresponding face database based on the obtained face feature vector and obtaining a corresponding similarity coefficient; A similarity threshold is set, and the obtained similarity coefficient is compared with the similarity threshold; and corresponding supervision instructions are generated according to the comparison result, and the supervision instructions include risk instructions and safety identification instructions.

7. A real-time property safety monitoring system for smart cities according to claim 5, characterized in that: The process of taking corresponding supervisory measures based on the risk instructions obtained includes: If the supervision instruction is a risk instruction, the real-time position of the corresponding person to be identified in the corresponding property community is obtained based on the obtained video frame image, and it is fed back to the monitoring center. The monitoring center takes the real-time position of the corresponding person to be identified as the center, notifies the nearest security personnel to patrol the corresponding person to be identified, and chooses whether to expel the corresponding person to be identified or call the police based on the patrol results.

8. A real-time property safety monitoring system for smart cities according to claim 5, characterized in that: The process of taking appropriate measures based on the security identification instructions obtained includes: If the corresponding supervision instruction is a security identification instruction, obtain the facial feature vector of the corresponding suspicious person; and based on it, search the video information collected by the corresponding monitoring terminal, and according to the search result, collect the corresponding information to the monitoring terminal related to the corresponding suspicious person, and mark it as an identification terminal; Obtain the real-time location and monitoring range of all identification terminals, and based on them, obtain the visit area ratio, security level and stay time corresponding to the corresponding suspicious persons; Conduct risk assessment on the corresponding suspicious persons based on the obtained visit area ratio, security level and stay time, and obtain corresponding risk scores; Set a risk threshold range and compare the obtained risk score with the risk threshold; If the risk score is lower than the minimum value of the risk threshold range, no other actions are performed; If the risk score is within the risk threshold range, the real-time location of the corresponding suspicious person is obtained in real time based on the monitoring terminal, and security patrols in the corresponding area are strengthened; If the risk score is higher than the maximum value of the risk threshold range, a risk warning will be fed back to the monitoring center, and the corresponding monitoring center will assign security personnel to verify the identity and patrol the corresponding suspicious persons.

Citation Information

Patent Citations

  • Data sharing method for smart community

    CN112002052A

  • Intelligent video monitoring alarm management system

    CN116343419A

  • Object tracking method and system based on face recognition and medium

    CN118522058A

  • Early warning method and system for regional public security management in smart city based on the internet of things

    US20230066101A1