Security and protection system based on user behavior analysis
Through a security system based on user behavior analysis, combined with face recognition and security analysis, the risk level is assessed and alarms are triggered, the problem of high resource waste and false alarm rates in the existing technology is solved, and abnormal ideas are achieved for users Precision detection and security improvement for strangers.
Patent Information
- Application Number
- CN202411044549.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-11-29
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing facial recognition security system is difficult to accurately detect strangers who have abnormal ideas on users, resulting in high resource waste and false alarm rates.
Design a security system based on user behavior analysis, and analyze user behavior and face through camera devices, facial reading recognition units, user identification units, user security analysis units, tracking units and alarm units, combined with face detection and recognition algorithms Expression, rating the risk level and triggering alarms, distinguishing users and passers-by, and reducing attention to passers-by and resource allocation.
It realizes accurate detection of strangers with abnormal ideas for users, reduces system resource consumption and false alarm rates, improves detection capabilities of malicious strangers, and ensures security and efficiency.
Smart Images

Figure CN119049104A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of image recognition, and in particular to a security system based on user behavior analysis. Background Art
[0002] With the development and progress of science and technology and the rapid development of information technology in the 21st century, intelligent security technology has entered a new field. The boundary between intelligent security technology and computers is gradually disappearing. Without security technology, society will be restless, and the progress and development of world science and technology will be affected. In recent years, with the rapid development of related technologies and the growing actual needs, face recognition has shown broad application prospects in many fields. For example, face recognition technology can be used as an auxiliary means in the judicial field for identity authentication, criminal identification, etc.; there are also many potential applications in business, such as bank credit card identity recognition, security identification system, videophone, medicine, archives management, hotel management, video conference human-computer interaction system, etc., which not only improves the efficiency of social operation, but also greatly enhances the safety of citizens' daily life. Therefore, face recognition technology has become a research hotspot in the current field of pattern recognition and artificial intelligence. Compared with other biometric recognition systems such as fingerprints, irises, genes, and voices, faces are popular because they are non-replicable, easy to collect, and do not require the cooperation of the person being photographed.
[0003] However, the current security system processing mechanism for face recognition is to identify whether the person photographing is a stranger, and then focus on detecting strangers. However, the majority of strangers are passers-by, while strangers who commit crimes are very rare. Therefore, a single centralized monitoring of strangers will only waste system energy and it is difficult to detect strangers who have abnormal thoughts about users. Summary of the invention
[0004] In view of the deficiencies of the prior art, the present invention provides a security system based on user behavior analysis, which has the advantages of being able to effectively screen targets and reduce system resource consumption, accurately detect strangers who have abnormal thoughts about users, and solves the problems of the above-mentioned technologies.
[0005] To achieve the above object, the present invention provides the following technical solutions: a security system based on user behavior analysis, the security system comprising a camera device for photographing a user's face, the camera device being built into a processor connected to a security system network, the security system further comprising a face reading and recognition unit, a user recognition unit, a user safety analysis unit, a tracking unit, and an alarm unit;
[0006] The facial reading and recognition unit is connected to the security system and is used to read the facial photo of the person being photographed taken by the camera device and recognize whether the face of the person being photographed is a user registered in the security system;
[0007] The user identification unit is connected to the security system and is used to identify whether the face of the person being photographed is a user, and if so, to mark the user, and if not, to mark the passerby;
[0008] The user safety analysis unit is connected to the user identification unit and is used to analyze the number of people being photographed. If the number is greater than one and both a passerby (l) and a user (y) appear, facial expressions of the passerby (l) and the user (y) are analyzed. If the number is greater than one and all passersby (l) appear, facial expression analysis of the passerby (l) is performed. If the number is greater than one and all users (y) appear, the system is not started. If the user (y) safety analysis unit analyzes that the user's facial expression shows negative emotions, the user is marked as a suspicious target and the tracking unit is started.
[0009] The tracking unit is used to periodically start the camera to shoot suspicious targets. The suspicious targets update their danger level every minute. The danger level is assessed as negative emotion +1, body movement +3 and the number of people is greater than one +3. If the above conditions are triggered simultaneously or individually, and the danger level is greater than 5, the alarm unit is activated to inform the security guard.
[0010] Preferably, the security system is also provided with a user entry unit, and the user entry unit is network-connected to a camera device and a personal mobile Internet terminal, the camera device and the personal mobile Internet terminal are used to take facial photos, and the user entry unit is used to enter the taken facial photos as users, and the user entry unit and the security system are in a two-way data transmission mode.
[0011] Preferably, the facial reading and recognition unit includes the following detailed steps:
[0012] S1.1. The image captured by the camera device is first processed by a face detection algorithm to determine whether there is a face in the image and extract the face part;
[0013] S1.2, the facial reading and recognition unit processes the extracted facial image using a facial recognition algorithm, and the processing includes facial alignment, feature extraction and feature matching;
[0014] S1.3, the security system compares the recognized facial features with the user information stored in the user information database in advance;
[0015] S1.4. Determine user identity: Based on the comparison results, the system can determine whether the face of the person being photographed is a user that has been entered into the system. If the match is successful, the identity of the person being photographed is confirmed to be the user (y); if it fails, the stranger's information is recorded and marked as pending information.
[0016] Preferably, the user identification unit further includes the following logical steps:
[0017] S2.1. The user identification unit receives pending information from the security system, where the pending information is facial information of the person being photographed;
[0018] S2.2, further processing and comparing the extracted facial features of the pending information through the face recognition algorithm;
[0019] S2.3. The extracted facial features are compared with the user information stored in the system to determine whether the person being photographed is a user that has been entered into the system. If the match is successful, the person being photographed is marked as a user (y); if the match fails, the person being photographed is marked as a passerby (l).
[0020] Preferably, the user identification unit expression is:
[0021] The input image is I, the face image extracted by the face detection algorithm is Face(I), the feature processed by the face recognition algorithm is Feature(Face(I)), and the user information stored in the user information database is User;
[0022] The matches that succeed are:
[0023] Match(Feature(Face(I)),User)=1;
[0024] The matching fails when:
[0025] Match(Feature(Face(I)),User)=0.
[0026] Preferably, the overall matching expression is:
[0027]
[0028] Preferably, the principle formula of the household security analysis unit is:
[0029]
[0030] in:
[0031] A stands for Analysis, which indicates the results of facial expression analysis in the presence of people, passers-by, and users;
[0032] F is the FacialExpressionAnalysis function that performs facial expression analysis, and the numbers in the brackets are passersby (l), users (y), or passersby and users;
[0033] NumOfPeople indicates the number of people photographed;
[0034] Exist(l) indicates the existence of passers-by;
[0035] Exist(y) means the user exists;
[0036] All(l) means all are passers-by;
[0037] All(y) means all are users.
[0038] Preferably, the tracking unit further includes the following detailed steps:
[0039] S3.1. The tracking unit receives the suspicious target information marked by the user security analysis unit;
[0040] S3.2, Regular shooting: This can be achieved by setting a timer or trigger condition, taking a picture every minute;
[0041] S3.3, Danger level update: After the shooting is completed, the danger level of the suspicious target is assessed and the danger level of the suspicious target is calculated according to the prescribed assessment standards;
[0042] S3.4, Alarm triggering condition check: Check whether the danger level of the suspected target meets the alarm triggering conditions;
[0043] S3.5. Alarm triggering: When the alarm condition is triggered, the tracking unit transmits the information to the alarm unit, which informs the security guard.
[0044] Preferably, the tracking unit expression is:
[0045] Assume that the motion of the tracking unit is represented by the vector To describe, where x(t) and y(t) represent the position of the tracking unit in the horizontal and vertical directions as a function of time, respectively, and the movement of the unit is represented by differentials assuming a constant velocity:
[0046]
[0047] Preferably, the security system further includes the following detailed steps:
[0048] Input data acquisition: obtain the captured image from the camera device, and extract the face image through the face detection algorithm to determine whether there is a face in the image and extract the face part;
[0049] Step 1: Face recognition: The extracted face image is passed to the face reading and recognition unit, and the face image is processed using a face recognition algorithm;
[0050] Step 2: User identity determination: comparing the recognized facial features with the user information stored in the user information database in advance;
[0051] Step 3: Processing pending information: The user identification unit receives pending information from the security system, i.e., facial information of the person being photographed;
[0052] Step 4: Secondary comparison: perform a secondary comparison between the extracted facial features and the user information stored in the system to determine whether the person being photographed is a user that has been entered into the system; if the match is successful, the person being photographed is marked as a user (y); if the match fails, the person being photographed is marked as a passerby (l);
[0053] Step 5: Safety analysis: Perform safety analysis based on the number of people photographed and the recognition results. During the analysis, if a user's facial expression shows negative emotions, it will be marked as a suspicious target and the tracking unit will be activated;
[0054] Step 6. Alarm triggering conditions: Based on the danger level assessment of the suspected target, the danger level is updated every minute; the danger level assessment is negative emotion +1, body movement +3 and the number of people is greater than one +3; if the danger level is greater than 5, the alarm unit is triggered and the alarm unit informs the security guard.
[0055] Compared with the prior art, the present invention provides a security system based on user behavior analysis, which has the following beneficial effects:
[0056] 1. The present invention analyzes and identifies user behavior. The system distinguishes the photographed persons into users and passers-by through facial recognition and user identity judgment, thereby performing targeted security analysis and reducing attention to passers-by and resource allocation. During the analysis process, the system comprehensively considers the number of people, users and passers-by, and only starts the tracking unit under specific conditions, thereby avoiding tracking and analyzing every photographed person and effectively reducing the consumption of system resources. The system triggers an alarm based on the danger level assessment of the suspicious target, rather than alarming every photographed person. This ensures the accuracy of the alarm, while reducing the false alarm rate, further saving system resources, and achieving the beneficial effect of being able to effectively screen targets and reduce system resource consumption.
[0057] 2. The present invention achieves advantages such as accurate detection of strangers with malicious intent to users through multi-level user behavior analysis and intelligent identification, thereby solving a series of technical challenges and bringing beneficial effects. The system can accurately distinguish between users and strangers through facial recognition and user identity judgment, ensuring that only registered users can be confirmed as legal entrants, effectively avoiding potential threats from strangers. The system performs security analysis based on the number of people, users and passers-by, and can promptly detect malicious behavior or potential threats through analysis of user facial expressions, thereby tracking and monitoring suspicious targets. The system triggers an alarm based on the danger level assessment of the suspicious target, and can respond quickly and accurately to potential risks, avoiding interference with normal users. The security system effectively improves the ability to detect malicious strangers through intelligent user behavior analysis and accurate target identification, achieves the advantages of accurate detection of strangers with abnormal intent to users, and solves the beneficial effects of the above-mentioned technologies. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] Figure 1 It is a schematic diagram of the system flow of the present invention; DETAILED DESCRIPTION
[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0060] See also Figure 1 , a security system based on user behavior analysis, the security system comprising a camera device for photographing a user's face, the camera device being built into a processor connected to a security system network, the security system further comprising a face reading and recognition unit, a user recognition unit, a user safety analysis unit, a tracking unit and an alarm unit;
[0061] The facial reading and recognition unit is connected to the security system and is used to read the facial photo of the person being photographed taken by the camera device and recognize whether the face of the person being photographed is a user registered in the security system;
[0062] The user identification unit is connected to the security system and is used to identify whether the face of the person being photographed is a user, and if so, to mark the user, and if not, to mark the passerby;
[0063] The user safety analysis unit is connected to the user identification unit and is used to analyze the number of people being photographed. If the number is greater than one and both a passerby (l) and a user (y) appear, facial expressions of the passerby (l) and the user (y) are analyzed. If the number is greater than one and all passersby (l) appear, facial expression analysis of the passerby (l) is performed. If the number is greater than one and all users (y) appear, the system is not started. If the user (y) safety analysis unit analyzes that the user's facial expression shows negative emotions, the user is marked as a suspicious target and the tracking unit is started.
[0064] The tracking unit is used to periodically start the camera to shoot suspicious targets. The suspicious targets update their danger level every minute. The danger level is assessed as negative emotion +1, body movement +3 and the number of people is greater than one +3. If the above conditions are triggered simultaneously or individually, and the danger level is greater than 5, the alarm unit is activated to inform the security guard.
[0065] Through facial recognition and user identity judgment, users and passers-by can be accurately identified, thus avoiding misjudgment and missed judgment, and improving the accuracy of the system. The user safety analysis unit can intelligently analyze potential security risks according to the number of people and facial expressions of the photographed persons, and promptly discover possible threatening behaviors, thereby ensuring safety. By comprehensively considering the number of people, users and passers-by, the system can start the tracking unit under specific conditions to avoid tracking and analyzing every person photographed, effectively saving system resources and improving operating efficiency. The tracking unit can regularly photograph suspicious targets and update the danger level assessment. Once the danger level exceeds the set threshold, the alarm unit can be quickly activated to achieve a rapid response to potential risks and ensure the safety of users. The system can automatically perform face recognition, security analysis and alarm triggering, reducing the need for manual intervention and improving the system's automation and efficiency.
[0066] Preferably, the security system is also provided with a user entry unit, and the user entry unit is network-connected to a camera device and a personal mobile Internet terminal, the camera device and the personal mobile Internet terminal are used to take facial photos, and the user entry unit is used to enter the taken facial photos as users, and the user entry unit and the security system are in a two-way data transmission mode.
[0067] By taking facial photos through cameras and personal mobile Internet terminals, users can quickly complete registration and entry, which simplifies the user entry process. The user entry unit can transmit data in both directions and can update user information in real time, including identity authentication and permission management, making user management more flexible and efficient.
[0068] Preferably, the facial reading and recognition unit includes the following detailed steps:
[0069] S1.1. The image captured by the camera device is first processed by a face detection algorithm to determine whether there is a face in the image and extract the face part;
[0070] S1.2, the facial reading and recognition unit processes the extracted facial image using a facial recognition algorithm, and the processing includes facial alignment, feature extraction and feature matching;
[0071] S1.3, the security system compares the recognized facial features with the user information stored in the user information database in advance;
[0072] S1.4. Determine user identity: Based on the comparison results, the system can determine whether the face of the person being photographed is a user that has been entered into the system. If the match is successful, the identity of the person being photographed is confirmed to be the user (y); if it fails, the stranger's information is recorded and marked as pending information.
[0073] Preferably, the user identification unit further comprises the following logical steps:
[0074] S2.1. The user identification unit receives pending information from the security system, where the pending information is facial information of the person being photographed;
[0075] S2.2, further processing and comparing the extracted facial features of the pending information through the face recognition algorithm;
[0076] S2.3. The extracted facial features are compared with the user information stored in the system to determine whether the person being photographed is a user that has been entered into the system. If the match is successful, the person being photographed is marked as a user (y); if the match fails, the person being photographed is marked as a passerby (l).
[0077] Preferably, the user identification unit expression is:
[0078] The input image is I, the face image extracted by the face detection algorithm is Face(I), the feature processed by the face recognition algorithm is Feature(Face(I)), and the user information stored in the user information database is User;
[0079] The matches that succeed are:
[0080] Match(Feature(Face(I)),User)=1;
[0081] The matching fails when:
[0082] Match(Feature(Face(I)),User)=0.
[0083] Preferably, the overall matching expression is:
[0084]
[0085] Using face detection and face recognition algorithms, facial features can be extracted and matched quickly and accurately, realizing efficient user identification function. Expressions provide clear classification of matching results, and different actions can be taken according to the matching results, enhancing the flexibility and applicability of the system.
[0086] Preferably, the principle formula of the household security analysis unit is:
[0087]
[0088] in:
[0089] A stands for Analysis, which indicates the results of facial expression analysis in the presence of people, passers-by, and users;
[0090] F is the FacialExpressionAnalysis function that performs facial expression analysis, and the numbers in the brackets are passersby (l), users (y), or passersby and users;
[0091] NumOfPeople indicates the number of people photographed;
[0092] Exist(l) indicates the existence of passers-by;
[0093] Exist(y) means the user exists;
[0094] All(l) means all are passers-by;
[0095] All(y) means all are users.
[0096] The formula takes into account the number of people being photographed, the presence of passers-by and users, and facial expression analysis in different situations, achieving a comprehensive analysis and evaluation of safety. Different facial expression analyses are performed according to the number of people being photographed and the presence of passers-by and users, to flexibly respond to possible safety risks in various situations, thereby enhancing the applicability and reliability of the system. The formula structure is clear and concise, easy to understand and implement, making the safety analysis logic clearer and easier to manage. Through the analysis of facial expressions, users who may have negative emotions can be identified in a timely manner and marked as suspicious targets, and the tracking unit can be started to achieve a rapid response to potential risks. The safety analysis process in the formula can be carried out automatically, which improves the system's automation level and efficiency and reduces the need for manual intervention.
[0097] Preferably, the tracking unit further includes the following detailed steps:
[0098] S3.1. The tracking unit receives the suspicious target information marked by the user security analysis unit;
[0099] S3.2, Regular shooting: This can be achieved by setting a timer or trigger condition, taking a picture every minute;
[0100] S3.3, Danger level update: After the shooting is completed, the danger level of the suspicious target is assessed and the danger level of the suspicious target is calculated according to the prescribed assessment standards;
[0101] S3.4, Alarm triggering condition check: Check whether the danger level of the suspected target meets the alarm triggering conditions;
[0102] S3.5. Alarm triggering: When the alarm condition is triggered, the tracking unit transmits the information to the alarm unit, which informs the security guard.
[0103] Preferably, the tracking unit expression is:
[0104] Assume that the motion of the tracking unit is represented by the vector To describe, where x(t) and y(t) represent the position of the tracking unit in the horizontal and vertical directions as a function of time, respectively, and the movement of the unit is represented by differentials assuming a constant velocity:
[0105]
[0106] Preferably, the security system further includes the following detailed steps:
[0107] Input data acquisition: obtain the captured image from the camera device, and extract the face image through the face detection algorithm to determine whether there is a face in the image and extract the face part;
[0108] Step 1: Face recognition: The extracted face image is passed to the face reading and recognition unit, and the face image is processed using a face recognition algorithm;
[0109] Step 2: User identity determination: comparing the recognized facial features with the user information stored in the user information database in advance;
[0110] Step 3: Processing pending information: The user identification unit receives pending information from the security system, i.e., facial information of the person being photographed;
[0111] Step 4: Secondary comparison: perform a secondary comparison between the extracted facial features and the user information stored in the system to determine whether the person being photographed is a user that has been entered into the system; if the match is successful, the person being photographed is marked as a user (y); if the match fails, the person being photographed is marked as a passerby (l);
[0112] Step 5: Safety analysis: Perform safety analysis based on the number of people photographed and the recognition results. During the analysis, if a user's facial expression shows negative emotions, it will be marked as a suspicious target and the tracking unit will be activated;
[0113] Step 6. Alarm triggering conditions: Based on the danger level assessment of the suspected target, the danger level is updated every minute; the danger level assessment is negative emotion +1, body movement +3 and the number of people is greater than one +3; if the danger level is greater than 5, the alarm unit is triggered and the alarm unit informs the security guard.
[0114] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and alterations may be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A security system based on user behavior analysis, the security system comprising a camera for photographing a user's face, the camera being built into a processor connected to a network of the security system, characterized in that: The security system also includes a facial reading and recognition unit, a user recognition unit, a user safety analysis unit, a tracking unit and an alarm unit; The facial reading and recognition unit is connected to the security system and is used to read the facial photo of the person being photographed taken by the camera device and recognize whether the face of the person being photographed is a user registered in the security system; The user identification unit is connected to the security system and is used to identify whether the face of the person being photographed is a user, and if so, to mark the user, and if not, to mark the passerby; The user safety analysis unit is connected to the user identification unit and is used to analyze the number of people being photographed. If the number is greater than one and both a passerby (l) and a user (y) appear, facial expressions of the passerby (l) and the user (y) are analyzed. If the number is greater than one and all passersby (l) appear, facial expression analysis of the passerby (l) is performed. If the number is greater than one and all users (y) appear, the system is not started. If the user (y) safety analysis unit analyzes that the user's facial expression shows negative emotions, the user is marked as a suspicious target and the tracking unit is started. The tracking unit is used to periodically start the camera to shoot suspicious targets. The suspicious targets update their danger level every minute. The danger level is assessed as negative emotion +1, body movement +3 and the number of people is greater than one +3. If the above conditions are triggered simultaneously or individually, and the danger level is greater than 5, the alarm unit is activated to inform the security guard.
2. A security system based on user behavior analysis according to claim 1, characterized in that: The security system is also provided with a user entry unit, which is network-connected to a camera device and a personal mobile Internet terminal. The camera device and the personal mobile Internet terminal are used to take facial photos. The user entry unit is used to enter the taken facial photos as users. The user entry unit and the security system are in a two-way data transmission mode.
3. A security system based on user behavior analysis according to claim 1, characterized in that: The facial recognition unit includes the following detailed steps: S1.
1. The image captured by the camera device is first processed by a face detection algorithm to determine whether there is a face in the image and extract the face part; S1.2, the facial reading and recognition unit processes the extracted facial image using a facial recognition algorithm, and the processing includes facial alignment, feature extraction and feature matching; S1.3, the security system compares the recognized facial features with the user information stored in the user information database in advance; S1.
4. Determine user identity: Based on the comparison results, the system can determine whether the face of the person being photographed is a user that has been entered into the system. If the match is successful, the identity of the person being photographed is confirmed to be the user (y); if it fails, the stranger's information is recorded and marked as pending information.
4. A security system based on user behavior analysis according to claim 1, characterized in that: The user identification unit also includes the following logical steps: S2.
1. The user identification unit receives pending information from the security system, where the pending information is facial information of the person being photographed; S2.2, further processing and comparing the extracted facial features of the pending information through the face recognition algorithm; S2.
3. The extracted facial features are compared with the user information stored in the system to determine whether the person being photographed is a user that has been entered into the system. If the match is successful, the person being photographed is marked as a user (y); if the match fails, the person being photographed is marked as a passerby (l).
5. A security system based on user behavior analysis according to claim 4, characterized in that: The user identification unit expression is: The input image is I, the face image extracted by the face detection algorithm is Face(I), the feature processed by the face recognition algorithm is Feature(Face(I)), and the user information stored in the user information database is User; The matches that succeed are: Match(Feature(face(I)),User)=1; The matching fails when: Match(Feature(Face(I)),User)=0.
6. A security system based on user behavior analysis according to claim 5, characterized in that: The overall matching expression is:
7. A security system based on user behavior analysis according to claim 6, characterized in that: The principle formula of the user safety analysis unit is: in: A stands for Analysis, which indicates the results of facial expression analysis in the presence of people, passers-by, and users; F is the FacialExpressionAnalysis function that performs facial expression analysis, and the numbers in the brackets are passersby (l), users (y), or passersby and users; NumOfPeople indicates the number of people photographed; Exist(l) indicates the existence of passers-by; Exist(y) means the user exists; All(l) means all are passers-by; All(y) means all are users.
8. A security system based on user behavior analysis according to claim 7, characterized in that: The tracking unit also includes the following detailed steps: S3.
1. The tracking unit receives the suspicious target information marked by the user security analysis unit; S3.2, Regular shooting: This can be achieved by setting a timer or trigger condition, taking a picture every minute; S3.3, Danger level update: After the shooting is completed, the danger level of the suspicious target is assessed and the danger level of the suspicious target is calculated according to the prescribed assessment standards; S3.4, Alarm triggering condition check: Check whether the danger level of the suspected target meets the alarm triggering conditions; S3.
5. Alarm triggering: When the alarm condition is triggered, the tracking unit transmits the information to the alarm unit, which informs the security guard.
9. A security system based on user behavior analysis according to claim 8, characterized in that: The tracking unit expression is: Assume that the motion of the tracking unit is represented by the vector To describe, where x(t) and y(t) represent the position of the tracking unit in the horizontal and vertical directions as a function of time, respectively, and the movement of the unit is represented by differentials assuming a constant velocity:
10. The security system based on user behavior analysis according to claim 1, characterized in that: The security system also includes the following detailed steps: Input data acquisition: obtain the captured image from the camera device, and extract the face image through the face detection algorithm to determine whether there is a face in the image and extract the face part; Step 1: Face recognition: The extracted face image is passed to the face reading and recognition unit, and the face image is processed using a face recognition algorithm; Step 2: User identity determination: comparing the recognized facial features with the user information stored in the user information database in advance; Step 3: Processing pending information: The user identification unit receives pending information from the security system, i.e., facial information of the person being photographed; Step 4: Secondary comparison: perform a secondary comparison between the extracted facial features and the user information stored in the system to determine whether the person being photographed is a user that has been entered into the system; if the match is successful, the person being photographed is marked as a user (y); if the match fails, the person being photographed is marked as a passerby (l); Step 5: Safety analysis: Perform safety analysis based on the number of people photographed and the recognition results. During the analysis, if a user's facial expression shows negative emotions, it will be marked as a suspicious target and the tracking unit will be activated; Step 6. Alarm triggering conditions: Based on the danger level assessment of the suspicious target, the danger level is updated every minute; the danger level assessment is negative emotion +1, body movement +3 and the number of people is greater than one +3; if the danger level is greater than 5, the alarm unit is triggered and the alarm unit informs the security guard.
Citation Information
Patent Citations
Intelligent door control safety method based on emotion recognition technology
CN111666780A
Public security video monitoring method and system based on face recognition
CN112183219A
Following identification method and computer readable storage medium
CN113284274A
