Operation control method for anti-theft safety door

Abstract

The present invention discloses an operation control method for an anti-theft security door, which relates to the technical field of anti-theft door control. The method comprises: when the security door is in a closed state, collecting the security door operation state through a sensor, wherein the security door operation state includes unlocking monitoring behavior and / or monitoring load behavior; performing abnormal analysis on the unlocking monitoring behavior and / or monitoring load behavior to generate a security door intrusion coefficient; when the security door intrusion coefficient is less than an intrusion coefficient threshold, activating an image collector, collecting a target person's portrait in front of the door, verifying the target person's portrait in front of the door through an authorized portrait library, and generating an identity verification result; when the identity verification result is not passed, or the security door intrusion coefficient is greater than or equal to the intrusion coefficient threshold, performing an audible and visual warning. The method solves the technical problem of high computing power cost caused by frequent portrait verification of security doors in the prior art, and achieves the technical effect of reducing the frequency of portrait verification and reducing the computing power occupancy rate of portrait verification.

Description

Technical Field

[0001] The present invention relates to the technical field of anti-theft door control, and in particular to an operation control method for an anti-theft safety door. Background Art

[0002] With the continuous advancement of technology and the increasing demand for public security, anti-theft security doors, as a vital component of modern security, have become widely used in various public places and important facilities. These doors not only effectively prevent unauthorized intrusion but also provide quick and effective authentication of those entering and exiting, thereby ensuring safety and order within the premises. Traditional anti-theft security doors typically use fingerprint recognition and facial verification to identify intruders. However, since each verification requires complex data processing and comparison, frequent fingerprint recognition and facial verification results in relatively high computing costs. This frequent verification further increases the system burden, especially in high-traffic environments. Summary of the Invention

[0003] The embodiments of the present application provide an operation control method for an anti-theft security door, which solves the technical problem in the prior art of high computing power cost caused by frequent portrait verification of security doors.

[0004] In view of the above problems, an embodiment of the present application provides an operation control method for an anti-theft security door.

[0005] An embodiment of the present application provides an operation control method for an anti-theft security door, the method comprising:

[0006] When the safety door is in a closed state, the operating state of the safety door is collected by a sensor, wherein the operating state of the safety door includes unlocking monitoring behavior and / or load monitoring behavior;

[0007] Performing abnormal analysis on the unlocking monitoring behavior and / or the monitoring load behavior to generate a safety door intrusion coefficient;

[0008] When the intrusion coefficient of the security door is less than the intrusion coefficient threshold, the image collector is activated to capture the target person's portrait in front of the door, and the target person's portrait in front of the door is verified through the authorized portrait library to generate an identity verification result;

[0009] When the identity verification result is failed, or the security door intrusion coefficient is greater than or equal to the intrusion coefficient threshold, an audible and visual warning is performed.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0011] When the security door is in a closed state, the operating status of the security door is collected through the sensor, wherein the operating status of the security door includes unlocking monitoring behavior and / or monitoring load behavior. Then, the unlocking monitoring behavior and / or monitoring load behavior are analyzed for abnormalities to generate a security door intrusion coefficient. When the security door intrusion coefficient is less than the intrusion coefficient threshold, the image collector is activated to collect the target person's portrait in front of the door, and the target person's portrait in front of the door is verified through the permission portrait library to generate an identity verification result; when the identity verification result is failed, or the security door intrusion coefficient is greater than or equal to the intrusion coefficient threshold, an audible and visual warning is performed. This solves the technical problem in the prior art that the computing power cost is high when the security door frequently performs portrait verification, and achieves the technical effect of reducing the frequency of portrait verification and reducing the computing power occupancy rate of portrait verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0013] Figure 1 A schematic flow chart of an operation control method for an anti-theft security door provided in an embodiment of the present application;

[0014] Figure 2 A schematic diagram of a flow chart for performing abnormality analysis in an operation control method for an anti-theft security door provided in an embodiment of the present application. DETAILED DESCRIPTION

[0015] The embodiments of the present application solve the technical problem in the prior art of high computing cost caused by frequent portrait verification of security doors by providing an operation control method for anti-theft security doors.

[0016] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0017] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0018] Example 1

[0019] like Figure 1 As shown, an embodiment of the present application provides an operation control method for an anti-theft security door, wherein the method includes:

[0020] When the safety door is in a closed state, the operating state of the safety door is collected by a sensor, wherein the operating state of the safety door includes unlocking monitoring behavior and / or load monitoring behavior;

[0021] When the safety door is closed, sensors monitor its operating status, including unlocking monitoring and / or load monitoring. Unlocking monitoring involves the system using sensors to monitor in real time whether the door is unlocked or attempting to unlock it. Load monitoring monitors whether the door is subject to abnormal external pressure or impact.

[0022] Furthermore, the operating status of the safety door is collected through sensors, including:

[0023] When the electronic password lock is in the unlock verification state, a list of unlock attempt passwords is collected and added to the unlock monitoring behavior;

[0024] The pressure vector time series information is collected through the pressure sensor and added to the monitoring load behavior.

[0025] Furthermore, the pressure sensor is deployed on the inner layer of the outer surface of the anti-theft safety door.

[0026] When the electronic combination lock is in the unlock verification state, the system collects a list of attempted unlock codes and adds it to the unlock monitoring behavior. The list of attempted unlock codes records all attempted unlock codes. A pressure sensor is deployed on the inner layer of the anti-theft security door's exterior surface to ensure that the sensor can accurately sense the external pressure applied to the security door while avoiding interference from external environmental factors. The pressure sensor collects pressure vector time series information and adds it to the monitoring load behavior. This pressure vector time series information reflects the pressure changes on the security door, including the magnitude, direction, and time series of the pressure changes.

[0027] Performing abnormal analysis on the unlocking monitoring behavior and / or the monitoring load behavior to generate a safety door intrusion coefficient;

[0028] Anomalies in unlocking monitoring behavior and / or monitoring load behavior are analyzed to generate a safety door intrusion coefficient, which reflects the intrusion risk or threat level currently faced by the safety door.

[0029] Furthermore, if Figure 2 As shown, performing abnormal analysis on the unlocking monitoring behavior and / or the monitoring load behavior to generate a safety door intrusion coefficient includes:

[0030] Performing deviation verification on the unlock monitoring behavior based on the unlock reference password to generate a target abnormality probability;

[0031] Processing the monitored load behavior through a behavior classifier to obtain a behavior classification result;

[0032] When the target abnormal probability is greater than or equal to the abnormal probability threshold, and / or the behavior classification result belongs to the intrusion behavior type, the security door intrusion coefficient is set to be greater than or equal to the intrusion coefficient threshold;

[0033] When the target abnormal probability is less than the abnormal probability threshold, and the behavior classification result does not belong to the intrusion behavior type, the security door intrusion coefficient is set to be less than the intrusion coefficient threshold.

[0034] Unlock monitoring behavior is verified for deviations against a preset unlock baseline password. The unlock baseline password is a valid, correct password used for comparison with attempted unlock passwords. By comparing the degree of deviation between each attempted password and the baseline password, a target anomaly probability is calculated. The target anomaly probability reflects the degree of inconsistency between the attempted password and the correct password, thus indicating the potential risk of unauthorized intrusion. A behavior classifier is used to process the monitored load behavior to obtain a behavior classification result. The behavior classifier is a pre-trained model that classifies pressure changes on the security door based on the time series information of the pressure vector collected by the pressure sensor, determining whether these pressure changes indicate normal operation, misoperation, or potential intrusion. The security door intrusion coefficient is updated based on the target anomaly probability and behavior classification results. If the target anomaly probability is greater than or equal to the preset anomaly probability threshold, or the behavior classification result indicates an intrusion behavior, the system will determine that there is a high intrusion risk and will set the security door intrusion coefficient to greater than or equal to the intrusion coefficient threshold. If the target anomaly probability is less than the anomaly probability threshold and the behavior classification result does not indicate an intrusion behavior, the system will determine that the current security risk is low and will set the security door intrusion coefficient to less than the intrusion coefficient threshold. Through detailed anomaly analysis, the system can more accurately determine whether the unlocking monitoring behavior and monitoring load behavior are abnormal, thereby generating an appropriate safety door intrusion coefficient.

[0035] Furthermore, the unlock monitoring behavior is subjected to deviation verification based on the unlock reference password to generate a target abnormality probability, including:

[0036] extracting the unlock attempt password list according to the unlock monitoring behavior;

[0037] Traversing the unlock attempt password list, calculating deviations from the unlock reference password, and generating a first deviation distance list;

[0038] Performing pairwise deviation calculation on the unlock attempt password list to generate a second deviation distance list;

[0039] An abnormal unlocking analysis is performed according to the first deviation distance list and the second deviation distance list to generate the target abnormality probability.

[0040] Optionally, based on the unlocking monitoring behavior, extract the unlocking attempt password list, which contains all the passwords that were attempted to unlock. Traverse the unlocking attempt password list, calculate the deviation between each attempted password and the unlocking reference password, and generate a first deviation distance list. The unlocking reference password is a preset correct password, which is used to compare with the attempted password. The first deviation distance list reflects the degree of deviation between each attempted password and the correct password. Perform pairwise deviation calculation on the unlocking attempt password list, that is, compare the deviation between any two attempted passwords in the list, and generate a second deviation distance list. The second deviation distance list contains the deviation information between the attempted passwords. Perform abnormal unlocking analysis based on the first deviation distance list and the second deviation distance list to generate the target abnormality probability.

[0041] Furthermore, the unlock attempt password list is traversed, deviations are calculated from the unlock reference password, and a first deviation distance list is generated, including:

[0042] Construct the deviation calculation function:

[0043] ;

[0044] ;

[0045] ;

[0046] in, The i-th unlock attempt password representing the unlock attempt password list, The j-th digit symbol representing the i-th unlock attempt password, Characterizes the unlocking base password, Represents the j-th digit symbol of the unlocking base password, M represents the number of password constraints, Characterizes the deviation distance between the i-th unlock attempt password and the unlock reference password;

[0047] According to the deviation calculation function, the unlock attempt password list is traversed, and deviation calculation is performed with the unlock reference password to generate the first deviation distance list.

[0048] The deviation calculation function uses the Euclidean distance formula to calculate the deviation between two passwords. In the deviation calculation function, The i-th unlock attempt password representing the unlock attempt password list The j-th digit symbol representing the i-th unlock attempt password, Unlock the base password. Represents the j-th digit symbol of the unlocking base password, M represents the number of password constraints, Characterize the deviation distance between the i-th unlock attempt password and the unlock reference password. According to the deviation calculation function, traverse the unlock attempt password list, calculate the deviation with the unlock reference password, and generate a first deviation distance list.

[0049] Furthermore, performing abnormal unlock analysis based on the first deviation distance list and the second deviation distance list to generate the target abnormality probability includes:

[0050] Calculating the variance of the first deviation distance list to generate a first dispersion coefficient;

[0051] Calculating the mean of the first deviation distance list to generate a first distance coefficient;

[0052] calculating the variance of the second deviation distance list to generate a second dispersion coefficient;

[0053] calculating a mean of the second deviation distance list to generate a second distance coefficient;

[0054] When the first dispersion coefficient is less than a first dispersion coefficient threshold, and the second dispersion coefficients are both less than a second dispersion coefficient threshold, and the first distance coefficient is less than a first distance coefficient threshold, and the second distance coefficient is less than a second distance coefficient threshold, the target abnormality probability is set to 0;

[0055] Otherwise, the target abnormality probability is set to 1.

[0056] The variance and mean of the first deviation distance list are calculated to generate a first dispersion coefficient and a first distance coefficient. The first dispersion coefficient reflects the degree of dispersion in the deviation distribution between the unlock attempt password and the baseline password, while the first distance coefficient reflects the average deviation between the unlock attempt password and the baseline password. The variance and mean of the second deviation distance list are calculated to generate a second dispersion coefficient and a second distance coefficient. If the first dispersion coefficient is less than the first dispersion coefficient threshold, the second dispersion coefficients are both less than the second dispersion coefficient threshold, the first distance coefficient is less than the first distance coefficient threshold, and the second distance coefficient is less than the second distance coefficient threshold—that is, if all four conditions are met, the unlock attempt password is considered to have a small deviation from the unlock baseline password as a whole, indicating no abnormal unlocking behavior. Therefore, the target abnormality probability is set to 0. If any of the conditions are not met, abnormal unlocking behavior is considered to have occurred, and the target abnormality probability is set to 1.

[0057] Furthermore, the monitored load behavior is processed by a behavior classifier to obtain a behavior classification result, including:

[0058] Configuring a pressure vector time series record data set and a behavior classification identification result, wherein the behavior classification identification result includes an intrusion behavior identification and a non-intrusion behavior identification;

[0059] Training a random forest based on the behavior classification identification result and the pressure vector time series record dataset, and generating the behavior classifier when the proportion of accurate classification times for a consecutive preset number of times is greater than or equal to a convergence proportion threshold;

[0060] The pressure vector time series information of the monitored load behavior is processed by a behavior classifier to obtain the behavior classification result.

[0061] The behavior classification identification results include intrusion behavior identification and non-intrusion behavior identification. The pressure vector time series record dataset is subjected to behavior classification identification, that is, the data in the pressure vector time series record dataset is divided into data with intrusion behavior identification and data with non-intrusion behavior identification based on historical experience. The behavior classification identification results are used as labels and the pressure vector time series record dataset is used as features to train the random forest model. During the training process, the number of accurate classifications for each iteration is recorded. When the proportion of accurate classifications for a consecutive preset number of times is greater than or equal to the convergence proportion threshold, the model is considered to have converged and a behavior classifier is generated. The pressure vector time series information of the monitored load behavior is input into the behavior classifier for processing to obtain the behavior classification results, thereby reflecting whether the current behavior is intrusion behavior or non-intrusion behavior.

[0062] When the intrusion coefficient of the security door is less than the intrusion coefficient threshold, the image collector is activated to capture the target person's portrait in front of the door, and the target person's portrait in front of the door is verified through the authorized portrait library to generate an identity verification result;

[0063] The calculated security door intrusion coefficient is compared with a preset threshold. When the coefficient falls below the threshold, the system activates an image collector, such as a camera, to capture the target person's image in front of the security door. After capturing the target person's image, the system verifies it against a permissioned portrait database, which contains facial images and relevant permission information for people allowed to enter the secure area. The captured target person's image is then compared with images in the permissioned portrait database to generate an identity verification result. This result indicates whether the target person has permission to open the door, thereby effectively monitoring and managing potential intrusions at the security door.

[0064] When the identity verification result is failed, or the security door intrusion coefficient is greater than or equal to the intrusion coefficient threshold, an audible and visual warning is performed.

[0065] The system will judge the identity verification result. If the identity of the target person does not match the information in the authorized portrait library, the identity verification result will be rejected. At the same time, the system will also check whether the security door intrusion coefficient is greater than or equal to the preset intrusion coefficient threshold. If the intrusion coefficient is greater than or equal to the intrusion coefficient threshold, it also indicates that an intrusion may have occurred in front of the security door. Regardless of whether the identity verification result is rejected or the security door intrusion coefficient is greater than or equal to the intrusion coefficient threshold, the system will immediately trigger an audible and visual warning. An audible and visual warning refers to the use of a high-decibel alarm sound through a siren or speaker to attract the attention of surrounding people and remind them that an intrusion may have occurred; a warning message is displayed through flashing lights or LED displays to visually remind people to pay attention to safety.

[0066] In summary, the embodiments of the present application have at least the following technical effects:

[0067] When the security door is in a closed state, the operating status of the security door is collected through the sensor, wherein the operating status of the security door includes unlocking monitoring behavior and / or monitoring load behavior. Then, the unlocking monitoring behavior and / or monitoring load behavior are analyzed for abnormalities to generate a security door intrusion coefficient. When the security door intrusion coefficient is less than the intrusion coefficient threshold, the image collector is activated to collect the target person's portrait in front of the door, and the target person's portrait in front of the door is verified through the permission portrait library to generate an identity verification result; when the identity verification result is failed, or the security door intrusion coefficient is greater than or equal to the intrusion coefficient threshold, an audible and visual warning is performed. This solves the technical problem in the prior art that the computing power cost is high when the security door frequently performs portrait verification, and achieves the technical effect of reducing the frequency of portrait verification and reducing the computing power occupancy rate of portrait verification.

[0068] It should be noted that the above-mentioned order of the embodiments of the present application is for descriptive purposes only and does not represent the superiority or inferiority of the embodiments. The above description is of specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0069] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0070] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. An operation control method for an anti-theft safety door, characterized in that: The method comprises: When the safety door is in a closed state, the operating state of the safety door is collected by a sensor, wherein the operating state of the safety door includes unlocking monitoring behavior and / or load monitoring behavior; Performing abnormal analysis on the unlocking monitoring behavior and / or the monitoring load behavior to generate a safety door intrusion coefficient; Performing deviation verification on the unlock monitoring behavior based on the unlock reference password to generate a target abnormality probability; Extracting a list of unlock attempt passwords based on the unlock monitoring behavior; Traversing the unlock attempt password list, calculating deviations from the unlock reference password, and generating a first deviation distance list; Performing pairwise deviation calculation on the unlock attempt password list to generate a second deviation distance list; Performing abnormal unlocking analysis based on the first deviation distance list and the second deviation distance list to generate the target abnormality probability; Processing the monitored load behavior through a behavior classifier to obtain a behavior classification result; Configuring a pressure vector time series record data set and a behavior classification identification result, wherein the behavior classification identification result includes an intrusion behavior identification and a non-intrusion behavior identification; Training a random forest based on the behavior classification identification result and the pressure vector time series record dataset, and generating the behavior classifier when the proportion of accurate classification times for a consecutive preset number of times is greater than or equal to a convergence proportion threshold; Processing the pressure vector time series information of the monitored load behavior through a behavior classifier to obtain the behavior classification result; When the target abnormal probability is greater than or equal to the abnormal probability threshold, and / or the behavior classification result belongs to the intrusion behavior type, the security door intrusion coefficient is set to be greater than or equal to the intrusion coefficient threshold; When the target abnormal probability is less than the abnormal probability threshold, and the behavior classification result does not belong to the intrusion behavior type, the security door intrusion coefficient is set to be less than the intrusion coefficient threshold; When the intrusion coefficient of the security door is less than the intrusion coefficient threshold, the image collector is activated to capture the target person's portrait in front of the door, and the target person's portrait in front of the door is verified through the authorized portrait library to generate an identity verification result; When the identity verification result is failed, or the security door intrusion coefficient is greater than or equal to the intrusion coefficient threshold, an audible and visual warning is performed.

2. The method according to claim 1, wherein The operating status of the safety door is collected through sensors, including: When the electronic password lock is in the unlock verification state, a list of unlock attempt passwords is collected and added to the unlock monitoring behavior; The pressure vector time series information is collected through the pressure sensor and added to the monitoring load behavior.

3. The method according to claim 2, wherein The pressure sensor is arranged on the inner layer of the outer surface of the anti-theft safety door.

4. The method according to claim 1, wherein Traversing the unlock attempt password list, calculating deviations from the unlock reference password, and generating a first deviation distance list, including: Construct the deviation calculation function: ； ； ； in, The i-th unlock attempt password representing the unlock attempt password list, The j-th digit symbol representing the i-th unlock attempt password, Characterizes the unlocking base password, Represents the j-th digit symbol of the unlocking base password, M represents the number of password constraints, Characterizes the deviation distance between the i-th unlock attempt password and the unlock reference password; According to the deviation calculation function, the unlock attempt password list is traversed, and deviation calculation is performed with the unlock reference password to generate the first deviation distance list.

5. The method according to claim 1, wherein Performing abnormal unlock analysis according to the first deviation distance list and the second deviation distance list to generate the target abnormality probability includes: Calculating the variance of the first deviation distance list to generate a first dispersion coefficient; Calculating the mean of the first deviation distance list to generate a first distance coefficient; calculating the variance of the second deviation distance list to generate a second dispersion coefficient; calculating a mean of the second deviation distance list to generate a second distance coefficient; When the first dispersion coefficient is less than a first dispersion coefficient threshold, and the second dispersion coefficients are both less than a second dispersion coefficient threshold, and the first distance coefficient is less than a first distance coefficient threshold, and the second distance coefficient is less than a second distance coefficient threshold, the target abnormality probability is set to 0; Otherwise, the target abnormality probability is set to 1.

Images