Intelligent security method and system based on face recognition

Abstract

The invention discloses an intelligent security and protection method and system based on face recognition, and the method comprises the steps: obtaining current parking data when a parking operation is received, analyzing the optimal detection radius of a current parking place for an authorized person in combination with historical parking data, and when the authorized person is detected to enter the optimal detection radius, carrying out the detection of the current parking place. Positioning information of the authorized personnel is obtained, the person whose contact ratio with the moving track of the authorized personnel is higher than a threshold value is identified as a peer according to the positioning information, a friendly value of the authorized personnel to the peer is analyzed, and different identities are given to the peer according to the friendly value. When it is detected that the authorized person searches the vehicle, whether the authorized person needs the vehicle searching navigation service or not is recognized according to the vehicle searching track of the authorized person, if yes, the vehicle searching navigation service is provided for the authorized person, and the vehicle searching navigation method and device have the advantages of improving system intelligence and optimizing user experience.

Description

Technical Field

[0001] The present invention relates to the technical field of face recognition, and specifically to an intelligent security method and system based on face recognition. Background Art

[0002] With the continuous progress of the intelligent vehicle industry, since the face recognition lock control system is in the unlocked state without anyone on duty for most of the time, and the parking environment is complex and changeable, it is necessary to adopt a 3D face recognition algorithm with a higher security level. With the help of dual-infrared camera technology, this algorithm can effectively adapt to application scenarios under various different environmental light conditions. Currently, some new models of cars have realized the car face recognition anti-theft system through a 3D binocular face recognition algorithm board, which not only effectively reduces the inconveniences caused by factors such as the car owner forgetting to bring the car key or signal jammers, but also improves the convenience and security of car use. However, the existing face recognition lock control system prohibits all unauthorized users from entering, resulting in a situation where the car owner's friend urgently needs to enter the car but the car owner is not there, so the car owner's friend is blocked outside the door, reducing the user experience. In addition, in the face of some underground garages with complex spatial layouts and dim light, the car owner often gets lost and cannot find his own car when coming and going, resulting in the car owner needing to spend a long time looking for the car, reducing the intelligence of the system. Therefore, it is very necessary to design an intelligent security method and system based on face recognition that improves the system intelligence and optimizes the user experience. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent security method and system based on face recognition to solve the problems raised in the above background art.

[0004] To solve the above technical problems, the present invention provides the following technical solution: An intelligent security method based on face recognition, the running steps of the method include:

[0005] Step S1: When receiving a parking operation, obtain the current parking data, and combine historical parking data to analyze the optimal detection radius for authorized personnel at the current parking location, where the optimal detection radius is the range threshold for locating the authorized personnel and triggering the car-finding navigation service;

[0006] Step S2: When it is detected that the authorized personnel enter the optimal detection radius, obtain the positioning information of the authorized personnel, identify those whose movement trajectories coincide with that of the authorized personnel with a coincidence degree higher than the threshold as accompanying personnel, analyze the friendliness value of the authorized personnel towards the accompanying personnel, and assign different identities to the accompanying personnel according to the friendliness value, and the identities include: authorized personnel and unauthorized personnel;

[0007] Step S3: When it is detected that the authorized person is looking for the vehicle, identify whether the authorized person needs the vehicle search navigation service according to the vehicle search trajectory of the authorized person. If so, provide the vehicle search navigation service to the authorized person.

[0008] Further, step S1 further includes the following steps:

[0009] Step S11: When it is detected that the parking operation is performed, construct an exclusive historical parking feature vector according to the historical parking data of different authorized persons using the vehicle search navigation.

[0010] Step S12: Construct an optimal detection radius model for the vehicle search navigation for the authorized person according to the historical parking feature vector, extract the parking feature vector of the current parking data, and input the parking feature vector into the optimal detection radius model to obtain the optimal detection radius of the vehicle search navigation for the authorized person at present.

[0011] Further, step S12 further includes the following steps:

[0012] Step S121: Identify whether there is a witness at the current parking location among the authorized persons. When there is a witness among the authorized persons, calculate the similarity between the current parking feature vector and the historical parking feature vector according to the historical parking feature vector of the witness, construct an optimal detection radius model for the vehicle search navigation according to the similarity and the historical optimal detection radius, and input the current parking feature vector into the optimal detection radius model to obtain the optimal detection radius. Herein, the witness refers to an authorized person who has seen the vehicle parked at the current parking location.

[0013] Step S122: When there is no such witness among the authorized persons, cluster the historical parking data of the authorized persons to generate scene categories, match them with the current parking data to obtain the scene category with the highest similarity, construct an optimal detection radius model for the vehicle search navigation according to the scene category, and input the current parking feature vector into the optimal detection radius model to obtain the optimal detection radius. Herein, the scene category refers to a grouped parking scene with similar features formed by clustering and analyzing the historical parking data of the authorized persons.

[0014] Further, step S2 further includes the following steps:

[0015] Step SA1: When it is detected that the authorized person enters within the corresponding optimal detection radius, match the facial data of the authorized person and identify the identity of the authorized person. Among the authorized persons, there are the first authorized person, the second authorized person, and the visitor authorized person. The first authorized person is a person who can use it for a long time and has the highest usage authority. The second authorized person can use it for a long time but has a lower usage authority than the first authorized person. The visitor authorized person is a temporary visitor of the vehicle and becomes an unauthorized person after the authorized time period.

[0016] Step SA2: Obtain the positioning accounts of the accompanying persons, match the positioning accounts with the positioning accounts in the historical authorization library. When it is identified that the accompanying persons include the unauthorized person, combine the surveillance video of the current parking location to grant the unauthorized person visitor status or retain the unauthorized person status.

[0017] Further, step SA2 further includes the following steps:

[0018] Step SA21: Obtain the facial data of the associated persons according to the positioning accounts, and match the facial data with the actual facial data in the surveillance video one by one. When the match is unsuccessful, trigger an abnormal alarm and switch the authorization status of the unmatched person. Among them, the positioning accounts between the facial data of the first authorized person and the second authorized person are associated.

[0019] Step SA22: When there is an unauthorized person among the accompanying persons, extract the facial micro-expressions and body movements of the accompanying persons and the authorized persons from the surveillance video, input the facial micro-expressions and body movements into the comprehensive body and face model of the authorized person and the unauthorized person, and obtain the friendliness value of the current authorized person towards the unauthorized person.

[0020] Step SA23: Compare the friendliness value with the first threshold. When the friendliness value is lower than the first threshold, turn on the in-vehicle video surveillance and the positioning function for the unauthorized person.

[0021] When the friendliness value is greater than the second threshold, grant the unauthorized person visitor authorization rights, where the first threshold is less than the second threshold.

[0022] Further, step SA23 further includes the following steps:

[0023] Step SA231: When the first authorized person is in the vehicle and an unauthorized person enters, extract the facial micro-expressions of the authorized person in the in-vehicle surveillance video to analyze the friendliness value towards the person, and judge whether to grant the unauthorized person visitor authorization rights according to the friendliness value.

[0024] Step SA232: When the first authorized person is not in the car, the in-car monitoring and positioning function is turned on and sent to the car owner's mobile phone in real time, so that the car owner can remotely authorize the identity of the target person through the mobile phone.

[0025] Furthermore, the system includes a parking navigation analysis module and an identity recognition and verification module:

[0026] The parking navigation analysis module is used to obtain current parking data when receiving a parking operation, and analyze the optimal detection radius of the current parking location for the authorized person in combination with historical parking data, wherein the optimal detection radius is a range threshold for locating the authorized person and triggering the car search navigation service;

[0027] The identity recognition and verification module is used to obtain the positioning information of the authorized person when it is detected that the authorized person enters the optimal detection radius, identify the person whose movement trajectory overlaps with the authorized person above a threshold as a companion person based on the positioning information, analyze the friendliness value of the authorized person to the companion person and assign different identities to the companion person based on the friendliness value, and the identities include: authorized personnel and unauthorized personnel.

[0028] Furthermore, the system also includes a behavior analysis response module:

[0029] The behavior analysis response module is used to identify whether the authorized person needs the car search navigation service based on the car search trajectory of the authorized person when it is detected that the authorized person is looking for a car, and if so, provide the car search navigation service to the authorized person.

[0030] The third aspect of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the electronic device implements the method described in the first aspect of the present application.

[0031] The fourth aspect of the present application provides a computer-readable storage medium, which is used to store a computer program. When the computer program runs on a computer, the computer executes the method described in the first aspect of the present application.

[0032] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: Through a multi-level dynamic permission management mechanism, combined with the fusion analysis of facial micro-expressions, body movements, and environmental data, the present invention significantly improves the intelligence and accuracy of vehicle safety protection. The system can not only quickly distinguish authorized personnel from unauthorized personnel based on face recognition but also dynamically adjust visitor permissions by real-time analyzing the emotional feedback of authorized personnel in the vehicle, effectively preventing security risks such as forced unlocking. At the same time, the innovative optimal detection radius model dynamically calculates through the adaptation of historical parking characteristics and environmental complexity, greatly improving the triggering accuracy and response efficiency of the car-finding navigation and reducing the interference of ineffective navigation push. For the in-depth trajectory analysis of the personnel's car-finding behavior, a multi-dimensional verification mechanism that integrates spatio-temporal features, physiological data, and environmental constraints is adopted. In addition, the cross-person navigation strategy optimization technology based on federated learning realizes the continuous evolution of the overall system performance under the premise of protecting privacy, solves the technical pain points of the rigid path planning and poor environmental adaptability of traditional navigation systems in complex parking lot environments, and thus improves the system intelligence and optimizes the personnel experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation to the present invention. In the drawings:

[0034] Figure 1 It is a schematic flowchart diagram of an intelligent security method based on face recognition provided by Embodiment 1 of the present invention.

[0035] Figure 2 It is a schematic diagram of the module composition of an intelligent security system based on face recognition provided by Embodiment 2 of the present invention.

[0036] Figure 3 It is a schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to 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 of 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.

[0038] This embodiment can be applied to the scenario of obtaining usage permissions through vehicle face recognition. This method can be executed by an intelligent security system based on face recognition provided in this embodiment. Figure 1 It is a schematic flowchart diagram of an intelligent security method based on face recognition provided by Embodiment 1 of the present invention. The method specifically includes the following steps:

[0039] Step S1: When receiving a parking operation, obtain the current parking data, and analyze the optimal detection radius for the authorized personnel at the current parking location in combination with historical parking data, where the optimal detection radius is the range threshold for locating the authorized personnel and triggering the car-finding navigation service;

[0040] Step S2: When it is detected that the authorized personnel enter the optimal detection radius, obtain the positioning information of the authorized personnel, identify the personnel whose movement trajectory coincides with that of the authorized personnel with a degree higher than the threshold as accompanying personnel, analyze the friendliness value of the authorized personnel towards the accompanying personnel, and assign different identities to the accompanying personnel according to the friendliness value, and the identities include: authorized personnel and unauthorized personnel;

[0041] Step S3: When it is detected that the authorized personnel are looking for their cars, identify whether the authorized personnel need the car-finding navigation service according to the car-finding trajectory of the authorized personnel. If so, provide the car-finding navigation service to the authorized personnel.

[0042] Specifically, through face recognition technology, the system can quickly and accurately identify the identities of target personnel, whether they are authorized or unauthorized personnel. This not only improves the security of vehicle access but also simplifies the personnel access process. Secondly, when the system receives a parking operation, it can intelligently analyze and determine the optimal detection radius in combination with historical data, which helps to obtain the positioning information of authorized personnel more accurately and provides more reliable data support for car-finding navigation. Finally, when the authorized personnel are looking for their cars, the system can intelligently judge whether car-finding navigation is needed according to their movement trajectory and send the navigation path in a timely manner when needed, which greatly improves the efficiency and accuracy of car-finding navigation and reduces the time and trouble for personnel to search for vehicles in the parking lot. Generally speaking, through intelligent face recognition, data analysis and behavior recognition technologies, this system realizes the automated and refined management of vehicle access and car-finding navigation, providing a safer, more convenient and efficient vehicle use experience for personnel.

[0043] In some preferred embodiments, step S1 further includes the following steps:

[0044] Step S11: When it is detected that the parking operation is performed, construct an exclusive historical parking feature vector according to the historical parking data of different authorized personnel using the car-finding navigation;

[0045] Step S12: Construct an optimal detection radius model for the car-finding navigation for the authorized personnel according to the historical parking feature vector, extract the parking feature vector of the current parking data, input the parking feature vector into the optimal detection radius model, and obtain the optimal detection radius of the authorized personnel for the current car-finding navigation.

[0046] Specifically, the optimization of the vehicle search process for authorized personnel in the parking lot is realized, significantly improving the accuracy and efficiency of vehicle search navigation. The historical parking data of authorized personnel is used to construct a historical parking feature vector, which provides important reference information for analyzing and predicting the current parking location. An optimal detection radius model is constructed based on the historical parking feature vector. This model can calculate the optimal detection radius according to the current parking location and the parking feature vector, thus providing more accurate positioning information for vehicle search navigation. The optimal detection radius is associated with the mobile positioning account of the authorized personnel to monitor the movement trajectory of the authorized personnel within the optimal detection radius in real time, ensuring that the vehicle search navigation service can respond to the needs of the authorized personnel in a timely and accurate manner. This process not only improves the efficiency of vehicle search navigation, reduces the time for personnel to search for vehicles in the parking lot, but also enhances the personnel experience, making the entire vehicle search process more convenient and intelligent.

[0047] In some preferred embodiments, the step S12 further includes the following steps:

[0048] Step S121: Identify whether there is a witness to the current parking location among the authorized personnel. When there is a witness among the authorized personnel, calculate the similarity s between the current parking feature vector and the historical parking feature vector according to the historical parking feature vector of the witness sim , construct the optimal detection radius model of the vehicle search navigation according to the similarity and the historical optimal detection radius, and input the current parking feature vector into the optimal detection radius model to obtain the optimal detection radius, where the witness refers to an authorized person who has seen the vehicle parked at the current parking location:

[0049]

[0050] In the formula, w k represents dynamic weight allocation, n represents the number of times the witness uses the vehicle search navigation, R witness represents the optimal detection radius of the witness for the current parking location, R hist (k) represents the historical optimal detection radius of the witness, γ witness and λ respectively represent the environmental and signal attenuation adjustment coefficients, C complex represents the complexity of the parking lot structure, S signal represents the signal strength of the positioning information;

[0051] Step S122: When there is no such witness among the authorized personnel, cluster the historical parking data of the authorized personnel to generate scene categories, match them with the current parking data to obtain the scene category with the highest similarity, construct an optimal detection radius model for the car-finding navigation according to the scene category, and input the current parking feature vector into the optimal detection radius model to obtain the optimal detection radius, where the scene category refers to a grouped parking scene with similar features formed after clustering and analyzing the historical parking data of the authorized personnel:

[0052]

[0053] In the formula, R witness represents the optimal detection radius of the authorized personnel for the current parking location, μ j represents the average value of the optimal detection radius within the scene category, and γ cluster represents the environmental adjustment coefficient.

[0054] Specifically, according to the complexity of the environmental structure of the historical parking location, analyze the optimal detection radius of different authorized personnel with different car-finding capabilities, and then accurately judge the car-finding intention of the authorized personnel, which can reduce unnecessary calculations and data processing, thereby reducing the energy consumption of the system, and identifying whether there is a witness among the authorized personnel can make the accuracy of the optimal detection radius more precise.

[0055] In some preferred embodiments, step S2 further includes the following steps:

[0056] Step SA1: When it is detected that the authorized personnel enter within their corresponding optimal detection radius, match the facial data of the authorized personnel and identify the authorized personnel. Among them, the authorized personnel include the first authorized personnel, the second authorized personnel, and the visitor authorized personnel. The first authorized personnel are those who can use it for a long time and have the highest usage rights. The second authorized personnel can be used for a long time but their usage rights are lower than those of the first authorized personnel. The visitor authorized personnel are temporary visitors of the vehicle and will be switched to unauthorized personnel after the authorized time period;

[0057] Step SA2: Obtain the positioning accounts of the accompanying personnel, match the positioning accounts with the positioning accounts in the historical authorization library. When it is identified that the accompanying personnel include the unauthorized personnel, combine the surveillance video of the current parking location to grant the unauthorized personnel visitor status or still retain the unauthorized personnel status.

[0058] Specifically, through precise face recognition technology and meticulous behavior analysis, rapid and accurate classification and identification of target personnel entering the monitoring range have been achieved. It covers the identification of first authorized personnel, second authorized personnel, and visitor authorized personnel, and also extends to the analysis of the facial micro-expressions and body movements of unauthorized personnel, thereby enabling a more detailed identity classification, improving the security and intelligence level of vehicle access control, while optimizing the personnel experience to ensure that only verified and authorized personnel can obtain access rights.

[0059] In some preferred embodiments, the step SA2 further includes the following steps:

[0060] Step SA21: Obtain the facial data of associated personnel according to the positioning account, and match the facial data one by one with the actual facial data in the surveillance video. When the match is unsuccessful, trigger an abnormal alarm and switch the authorized identity of the unmatched personnel. Among them, the positioning accounts between the facial data of the first authorized personnel and the second authorized personnel are associated;

[0061] Step SA22: When there is an unauthorized person among the accompanying personnel, extract the facial micro-expressions and body movements of the accompanying personnel and the authorized personnel from the surveillance video, input the facial micro-expressions and body movements into the comprehensive body and face model of the authorized personnel and the unauthorized personnel, and obtain the friendliness value of the current authorized personnel towards the unauthorized person;

[0062] Step SA23: Compare the friendliness value with a first threshold. When the friendliness value is lower than the first threshold, turn on the in-vehicle video monitoring and the positioning function for the unauthorized person;

[0063] When the friendliness value is greater than a second threshold, grant the unauthorized person the visitor authorization permission, where the first threshold is less than the second threshold.

[0064] Specifically, through high-precision face recognition technology and meticulous behavior analysis, intelligent identification and friendliness evaluation of unauthorized personnel among the accompanying personnel have been achieved. When the facial data associated with the positioning account does not match the actual facial data in the surveillance video, the system can quickly trigger an abnormal alarm and deprive the unauthorized identity of the unmatched personnel, effectively enhancing the security of vehicle access. At the same time, the system can also accurately evaluate the friendliness value between the two parties based on the comprehensive analysis of the facial micro-expressions and body movements of the authorized personnel towards the unauthorized person, and decide whether to grant the unauthorized person the visitor authorization permission accordingly.

[0065] In some alternative embodiments, the step SA22 further includes: The comprehensive body and face model of the authorized personnel and the unauthorized personnel is:

[0066]

[0067] Wherein, T face represents the facial emotion value of the authorized person, ω i represents the weight coefficient of the emotion of the i-th authorized person, E i represents the facial emotion value of the i-th authorized person, E base represents the daily expression baseline of the i-th authorized person, S threat represents the friendliness value of the current authorized person towards the unauthorized person, and α, β, and γ represent weight coefficients, A k represents the body movement detection value of the person in the k-th video frame, ω k represents the body weight coefficient of the person in the k-th video frame, N alert represents the historical body movements of the authorized person, t represents the time variable, and N represents the number of times the unauthorized person is given the temporary access right.

[0068] In some preferred embodiments, the step SA23 further includes the following steps:

[0069] Step SA231: When the first authorized person is in the vehicle and an unauthorized person enters, extract the facial micro-expression analysis of the authorized person in the in-vehicle surveillance video to obtain the friendliness value of the person, and determine whether to grant the unauthorized person visitor authorization rights according to the friendliness value;

[0070] Step SA232: When the first authorized person is not in the vehicle, turn on the in-vehicle surveillance and positioning functions and send them to the owner's mobile phone in real time, so that the owner can remotely authorize the identity of the target person through the mobile phone.

[0071] Specifically, a more flexible and personalized vehicle access control strategy. When there is a first authorized person in the vehicle and an unauthorized person appears, the system can use the in-vehicle surveillance video to capture and analyze the facial micro-expressions of the authorized person to accurately evaluate their friendliness towards the unauthorized person, and accordingly decide whether to grant visitor permissions. This not only improves the intelligence level of vehicle access but also ensures the harmony and safety of the in-vehicle environment. In the case where there is no first authorized person in the vehicle, the system automatically turns on the in-vehicle surveillance and positioning functions and sends real-time information to the owner's mobile phone, enabling the owner to remotely authorize the identity of the target person. This function greatly enhances the convenience and safety of vehicle use, allowing the owner to master and control the vehicle access rights at any time, achieving the integration of intelligence and personalization in vehicle management.

[0072] Based on the same inventive concept as the above method embodiment, the embodiment of the present invention further provides an intelligent security system based on face recognitionFigure 2 This is a schematic diagram of the module composition of an intelligent security system based on face recognition provided by an embodiment of the present invention. As Figure 2 shown, the system includes a parking navigation analysis module, an identity recognition and verification module, and a behavior analysis and response module:

[0073] The parking navigation analysis module is used to obtain current parking data when a parking operation is received, and combine historical parking data to analyze the optimal detection radius of the current parking location for authorized personnel. Among them, the optimal detection radius is the range threshold for locating the authorized personnel and triggering the car-finding navigation service;

[0074] The identity recognition and verification module is used to obtain the positioning information of the authorized personnel when it is detected that the authorized personnel enter the optimal detection radius. According to the positioning information, those whose movement trajectories coincide with that of the authorized personnel above a threshold are identified as accompanying personnel, and the friendliness value of the authorized personnel towards the accompanying personnel is analyzed, and different identities are assigned to the accompanying personnel according to the friendliness value. The identities include: authorized personnel and unauthorized personnel;

[0075] The behavior analysis and response module is used to identify whether the authorized personnel need the car-finding navigation service according to the car-finding trajectory of the authorized personnel when it is detected that the authorized personnel are looking for their cars. If so, provide the car-finding navigation service to the authorized personnel.

[0076] It should be noted that although several units or subunits of the device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present application, the features and functions of the two or more units described above can be embodied in one unit. Conversely, the features and functions of one unit described above can be further divided and embodied by multiple units.

[0077] Based on the same inventive concept as the above method embodiment, an embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the electronic device implements the control method in the above embodiment.

[0078] In one embodiment, the electronic device may be a server. In this embodiment, the structure of the electronic device may be as Figure 3 shown, including a memory 2001, a communication module 2003, and one or more processors 2002.

[0079] A memory 2001 for storing a computer program executed by a processor 2002. The memory 2001 may mainly include a program storage area and a data storage area. Among them, the program storage area may store an operating system and programs required to run an instant messaging function, etc.; the data storage area may store various instant messaging information and operation instruction sets, etc.

[0080] The memory 2001 may be a volatile memory, such as a random-access memory (RAM); the memory 2001 may also be a non-volatile memory, such as a read-only memory, a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD); or the memory 2001 is any other medium capable of carrying or storing a desired computer program in the form of instructions or data structures and accessible by a computer, but not limited thereto. The memory 2001 may be a combination of the above memories.

[0081] The processor 2002 may include one or more central processing units (CPUs) or be a digital processing unit, etc. The processor 2002 is used to implement the above audio data processing method when calling the computer program stored in the memory 2001.

[0082] A communication module 2003 is used to communicate with a terminal device and other servers.

[0083] In the embodiments of the present application, the specific connection medium between the above-mentioned memory 2001, communication module 2003, and processor 2002 is not limited. In the embodiments of the present application Figure 3 it is described that the memory 2001 and the processor 2002 are connected through a bus 2004, and the bus 2004 is described by an arrow in Figure 3 The connection methods between other components are only for illustrative purposes and are not to be construed as limiting. The bus 2004 may be divided into an address bus, a data bus, a control bus, etc. For ease of description, Figure 3 only one arrow is used to describe it in

[0084] Based on the same inventive concept as the above method embodiments, an embodiment of the present invention further provides a computer-readable storage medium, which is used to store a computer program. When the computer program runs on a computer, the electronic device is enabled to implement the control method in the above embodiments. The computer-readable storage medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0085] Based on the same inventive concept as the above method embodiments, an embodiment of the present invention further provides a computer program product, which includes a computer program. When the program product runs on an electronic device, the computer program is used to enable the electronic device to execute the steps in the control method according to various exemplary embodiments described above in this specification. The program product can adopt any combination of one or more readable media. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a means for realizing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 or a means for realizing the functions specified in multiple blocks.

[0086] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications falling within the scope of the present application.

Claims

1. An intelligent security method based on face recognition, characterized in that: When a parking operation is received, obtain the current parking data, and combine the historical parking data to analyze the optimal detection radius of the authorized personnel at the current parking location, where the optimal detection radius is the range threshold for locating the authorized personnel and triggering the car-finding navigation service; When it is detected that the authorized personnel enter the optimal detection radius, obtain the positioning information of the authorized personnel, identify the people traveling together whose movement trajectories coincide with that of the authorized personnel with a coincidence degree higher than the threshold according to the positioning information, analyze the friendliness value of the authorized personnel to the people traveling together, and assign different identities to the people traveling together according to the friendliness value, and the identities include: authorized personnel and unauthorized personnel; When it is detected that the authorized personnel are looking for their cars, identify whether the authorized personnel need the car-finding navigation service according to the car-finding trajectory of the authorized personnel. If so, provide the car-finding navigation service to the authorized personnel.

2. The intelligent security method based on face recognition according to claim 1, wherein: The step of obtaining the current parking data when a parking operation is received and combining the historical parking data to analyze the optimal detection radius of the authorized personnel at the current parking location includes: When the parking operation is detected, construct an exclusive historical parking feature vector according to the historical parking data of different authorized personnel using the car-finding navigation; Construct an optimal detection radius model for the car-finding navigation for the authorized personnel according to the historical parking feature vector, extract the parking feature vector of the current parking data, input the parking feature vector into the optimal detection radius model, and obtain the optimal detection radius of the authorized personnel for the current car-finding navigation.

3. The intelligent security method based on face recognition according to claim 2, wherein: The step of constructing an optimal detection radius model for the car-finding navigation for the authorized personnel according to the historical parking feature vector, extracting the parking feature vector of the current parking data, inputting the parking feature vector into the optimal detection radius model, and obtaining the optimal detection radius of the authorized personnel for the current car-finding navigation includes: Identify whether there is a witness to the current parking location among the authorized personnel. When there is a witness among the authorized personnel, calculate the similarity between the current parking feature vector and the historical parking feature vector according to the historical parking feature vector of the witness, construct an optimal detection radius model for the car-finding navigation according to the similarity and the historical optimal detection radius, and input the current parking feature vector into the optimal detection radius model to obtain the optimal detection radius, where the witness refers to an authorized person who has seen the car parked at the current parking location; When there is no witness among the authorized personnel, cluster the historical parking data of the authorized personnel to generate scene categories, match them with the current parking data to obtain the scene category with the highest similarity, construct an optimal detection radius model for the car-finding navigation according to the scene category, and input the current parking feature vector into the optimal detection radius model to obtain the optimal detection radius, where the scene category refers to a grouped parking scene with similar characteristics formed by clustering and analyzing the historical parking data of the authorized personnel.

4. An intelligent security method based on face recognition according to claim 3, characterized in that: When it is detected that the authorized person enters within the optimal detection radius, obtain the positioning information of the authorized person, identify the fellow travelers whose movement trajectories coincide highly with the threshold according to the positioning information, and analyze the friendliness value of the authorized person towards the fellow travelers and assign identities to the fellow travelers, including: When it is detected that the authorized person enters within the corresponding optimal detection radius, match the facial data of the authorized person and identify the identity of the authorized person. Among the authorized persons, there are the first authorized person, the second authorized person, and the visitor authorized person. The first authorized person is a person who can use it for a long time and has the highest usage authority. The second authorized person can be used for a long time but has a lower usage authority than the first authorized person. The visitor authorized person is a temporary visitor to the vehicle and becomes an unauthorized person after the authorized time period; Obtain the positioning account of the fellow traveler, match the positioning account with the positioning accounts in the historical authorization library. When it is identified that the fellow traveler includes the unauthorized person, in combination with the surveillance video at the current parking location, assign the unauthorized person the identity of a visitor or retain the identity of the unauthorized person.

5. The intelligent security method based on face recognition according to claim 4, wherein: The obtaining the positioning account of the fellow traveler, matching the positioning account with the positioning accounts in the historical authorization library. When it is identified that the fellow traveler includes the unauthorized person, the identity classification of the unauthorized person in combination with the surveillance video at the current parking location includes: Obtain the facial data of the associated person according to the positioning account, match the facial data with the actual facial data in the surveillance video one by one. When the match is unsuccessful, trigger an abnormal alarm and switch the authorization identity of the unmatched person. Among them, the positioning accounts between the facial data of the first authorized person and the second authorized person are associated; When there is an unauthorized person among the fellow travelers, extract the facial micro-expressions and body movements of the fellow travelers and the authorized person from the surveillance video, input the facial micro-expressions and body movements into the comprehensive body and face model of the authorized person and the unauthorized person, and obtain the current friendliness value of the authorized person towards the unauthorized person; Compare the friendliness value with the first threshold. When the friendliness value is lower than the first threshold, turn on the in-vehicle video surveillance and the positioning function for the unauthorized person; When the friendliness value is greater than the second threshold, grant the unauthorized person the visitor authorization permission, where the first threshold is less than the second threshold.

6. The intelligent security method based on face recognition according to claim 5, characterized in that: The comparing the friendliness value with the first threshold. When the friendliness value is lower than the first threshold, do not grant the unauthorized person the visitor authorization permission and turn on the in-vehicle video surveillance and the positioning function. When the friendliness value is higher than the first threshold and lower than the second threshold, do not grant the unauthorized person the visitor authorization permission. When the friendliness value is greater than the second threshold, grant the unauthorized person the visitor authorization permission, including: When there is the first authorized person in the vehicle and an unauthorized person enters, the facial micro-expressions of the authorized person in the in-vehicle surveillance video are extracted to analyze the friendliness value of the person, and it is determined whether to grant the unauthorized person the visitor authorization permission according to the friendliness value; When there is no first authorized person in the vehicle, the in-vehicle surveillance and positioning functions are activated and sent to the mobile phone of the vehicle owner in real time, so that the vehicle owner can remotely authorize the identity of the target person through the mobile phone.

7. An intelligent security system based on face recognition, characterized in that: The system includes a parking navigation analysis module and an identity recognition and verification module: The parking navigation analysis module is used to obtain the current parking data when a parking operation is received, and combine the historical parking data to analyze the optimal detection radius of the current parking location for the authorized person, where the optimal detection radius is the range threshold for locating the authorized person and triggering the car-finding navigation service; The identity recognition and verification module is used to obtain the positioning information of the authorized person when it is detected that the authorized person enters the optimal detection radius, identify the person whose movement trajectory coincides with that of the authorized person with a coincidence degree higher than the threshold as a companion, analyze the friendliness value of the authorized person towards the companion, and assign different identities to the companion according to the friendliness value. The identities include: authorized person and unauthorized person.

8. An intelligent security system based on face recognition according to claim 7, characterized in that: The system further includes a behavior analysis and response module: The behavior analysis and response module is used to identify whether the authorized person needs the car-finding navigation service according to the car-finding trajectory of the authorized person when it is detected that the authorized person is looking for the car. If so, the car-finding navigation service is provided to the authorized person.

9. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, the electronic device implements the method according to any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program. When the computer program runs on a computer, the computer executes the method according to any one of claims 1 to 6.

Images