Intelligent door lock management system

Abstract

The invention relates to the technical field of security door locks, in particular to an intelligent door lock management system which comprises a data acquisition module, an extraction module, a first determination module, a second determination module, a judgment module, an identification module, an adjustment module, a correction module and an alarm module. Through multi-level identity verification and abnormal behavior detection, the safety of the door lock is improved, the key pressure, the sliding track and the face image are acquired in real time through the data acquisition module, the face features and the shielding rate are extracted in combination with the extraction module, preliminary recognition of the target identity is achieved, and the safety of the door lock is improved. The key pressure and track consistency degree is calculated through a second determination module, whether an abnormal door opening mode exists or not is judged, a judgment module integrates the abnormal mode and the shielding rate and confirms the target identity and permission, and a recognition module recognizes the permission value according to the permission library and conducts dynamic adjustment. The problem that the prevention effect on complex security threats is low due to the fact that recognition depends on single password input is effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of security door locks, and in particular to an intelligent door lock management system. Background Art

[0002] With the popularization of smart home technology, access control systems in places such as classrooms and dormitories are gradually adopting smart door locks to improve security and convenience. However, existing smart door lock systems often rely on traditional password input, fingerprint recognition or card unlocking in these scenarios, which have security risks such as password leakage and recognition errors, and lack effective monitoring and real-time response to abnormal behavior. Especially when multiple people use or the environment is complex, the existing system is difficult to intelligently identify potential abnormal unlocking behaviors, which increases security risks. Therefore, a smarter and safer door lock management system is urgently needed to meet these challenges.

[0003] The patent document with patent publication number CN107341885A discloses an intelligent door lock management system, which includes: a door lock device, a mobile terminal and a cloud platform, the door lock device includes a communication module, a control module, an electronic lock and a keyboard controller; the communication module, the electronic lock and the keyboard controller are respectively connected to the control module, the communication module is connected to the cloud platform by wireless or wired means, and the cloud platform is connected to the mobile terminal by wireless or wired means; the communication module is used to receive commands sent by the mobile terminal and / or the cloud platform, and transmit them to the control module; the keyboard controller is used to input the electronic lock password; the control module is used to parse the received commands and / or password data, and control the unlocking state of the electronic lock and the password controller manages the unlocking password of the electronic lock; the electronic lock and the control module interact with each other for the electronic lock unlocking state data, so that the control module can feedback the state of the electronic lock to the mobile terminal and / or the cloud platform.

[0004] It can be seen that the smart door lock management system has the following problems: the door lock device of the system mainly relies on the keyboard controller to input the password, which limits the system's ability to identify abnormal behavior and security; the system mainly parses password data and executes commands through the control module, and the judgment mechanism for abnormal door opening behavior is weak, and it is unable to actively identify and warn of possible security threats; the system relies on static data, and is unable to adjust the judgment parameters according to actual conditions during long-term use, and there is a potential risk of misjudgment or missed judgment. Summary of the invention

[0005] To this end, the present invention provides an intelligent door lock management system, which is used to overcome the problem of low prevention effect against complex security threats in the prior art due to reliance on single password input recognition through multimodal data collection and intelligent judgment mechanism.

[0006] To achieve the above object, the present invention provides an intelligent door lock management system, comprising:

[0007] A data acquisition module is used to collect the real-time key pressure and sliding trajectory of the target to be identified at the door lock during the door opening process, as well as the real-time image of the target to be identified;

[0008] An extraction module, used for extracting real-time facial feature values ​​and real-time facial occlusion rate of the real-time image;

[0009] A first determination module, connected to the extraction module, for determining a first target according to the real-time facial feature value and a preset comparison feature value;

[0010] a second determination module, which is connected to the data acquisition module and the first determination module respectively, and is used to determine an abnormal door opening mode according to the real-time key pressure and the real-time sliding trajectory of the first target;

[0011] a determination module, which is connected to the second determination module and the extraction module respectively, and is used to determine the second target according to the number of the abnormal door opening patterns and the real-time facial occlusion rate within a preset determination time;

[0012] an identification module, which is connected to the first determination module and the judgment module respectively, and is used to identify the authority value of the first target according to a preset authority library to form a first authority value, and to identify the authority value of the second target to form a second authority value;

[0013] an adjustment module, connected to the identification module and the first determination module respectively, for adjusting the preset comparison feature value according to the first authority value and the second authority value to form an adjusted feature value;

[0014] a correction module, which is connected to the adjustment module and the second determination module respectively, and is used to correct the preset determination time length according to the number of the adjustment characteristic values ​​within the preset correction time length;

[0015] An alarm module is connected to the determination module and is used to issue an intrusion alarm according to the second target determined based on the modified preset determination time.

[0016] Furthermore, the first determining module includes:

[0017] A feature comparison unit, used to compare the real-time facial feature value with the preset comparison feature value to form a feature comparison result;

[0018] A first determination unit is connected to the feature comparison unit, and is used to determine that the target to be identified is a first target when the feature comparison result is that the real-time facial feature value is greater than the preset comparison feature value.

[0019] Furthermore, the second determining module includes:

[0020] A first calculation unit, configured to calculate a pressure consistency according to the real-time key pressure and the historical key pressure within a preset historical time period;

[0021] A second calculation unit is used to calculate the trajectory consistency according to the real-time sliding trajectory and the historical sliding trajectory within the preset historical time period;

[0022] A second determining unit is connected to the first calculating unit and the second calculating unit respectively, and is used to determine an abnormal door opening mode according to the pressure consistency and the trajectory consistency.

[0023] Furthermore, the first computing unit includes:

[0024] A historical pressure calculation subunit, used to calculate an average value of historical key pressures within a preset historical time period to form an average historical pressure, and to calculate a standard deviation of historical key pressures within a preset historical time period to form a historical pressure standard deviation;

[0025] The pressure consistency calculation subunit is connected to the historical pressure calculation subunit, and is used to calculate the absolute value of the difference between the real-time key pressure and the average historical pressure, and to calculate the ratio of the absolute value to the historical pressure standard deviation to form pressure consistency.

[0026] Furthermore, the second computing unit includes:

[0027] The historical trajectory point calculation subunit is used to calculate the average value of the sliding trajectory points at each time point within a preset historical time length to form a number of average historical trajectory points;

[0028] A trajectory drawing subunit, connected to the historical trajectory point calculation subunit, for drawing an average historical sliding trajectory according to the average historical trajectory points;

[0029] The trajectory consistency calculation subunit is connected to the trajectory drawing subunit and is used to calculate the cosine similarity between the average historical sliding trajectory and the real-time sliding trajectory to form the trajectory consistency.

[0030] Further, the second determining unit includes:

[0031] A pressure comparison subunit, used to compare the pressure consistency with a preset pressure consistency threshold to form a pressure comparison result;

[0032] A trajectory comparison subunit, used to compare the trajectory consistency with a preset trajectory consistency threshold to form a trajectory comparison result;

[0033] The second determination subunit is connected to the pressure comparison subunit and the trajectory comparison subunit respectively, and is used to determine that an abnormal door opening mode occurs when the pressure comparison result is that the pressure consistency is less than a preset pressure consistency threshold, and the trajectory comparison result is that the trajectory consistency is less than a preset trajectory consistency threshold.

[0034] Furthermore, the determination module includes:

[0035] An abnormality comparison unit, used to compare the number of abnormal door opening modes with a preset abnormal number threshold to form an abnormality comparison result;

[0036] An occlusion comparison unit, used to compare the real-time facial occlusion rate with a preset occlusion rate threshold to form an occlusion comparison result;

[0037] A determination unit is connected to the abnormal comparison unit and the occlusion comparison unit, respectively, and is used to determine that the first target is the second target when the abnormal comparison result is that the number of the abnormal door opening patterns is less than the preset abnormal number threshold, and the real-time facial occlusion rate is greater than the preset occlusion rate threshold.

[0038] Furthermore, the determination unit is also used to determine that the target to be identified has no access permission when the abnormal comparison result is that the number of the abnormal door opening patterns is greater than the preset abnormal number threshold.

[0039] Furthermore, the adjustment module includes:

[0040] a ratio calculation unit, used to calculate a relative deviation between the first authority value and the second authority value to form an adjustment ratio;

[0041] a ratio comparison unit, connected to the ratio calculation unit, for comparing the adjustment ratio with a preset ratio threshold to form a ratio comparison result;

[0042] The adjustment unit is connected to the ratio comparison unit, and is used to increase the preset comparison characteristic value according to the adjustment ratio and a preset adjustment coefficient to form an adjustment characteristic value when the ratio comparison result is that the adjustment ratio is greater than the preset ratio threshold.

[0043] Furthermore, the correction module includes:

[0044] A fluctuation calculation unit, used to calculate the standard deviation of the quantity of the adjusted characteristic value to form a quantity fluctuation value;

[0045] A correction unit is connected to the fluctuation calculation unit and is used to increase the preset determination time according to a relative deviation between the quantity fluctuation value and the preset fluctuation value threshold and a preset correction coefficient when the quantity fluctuation value is greater than the preset fluctuation value threshold.

[0046] Compared with the prior art, the beneficial effect of the present invention is that the security of the door lock is improved through multi-level identity authentication and abnormal behavior detection. The key pressure, sliding trajectory and facial image are obtained in real time through the data acquisition module, and the facial features and occlusion rate are extracted in combination with the extraction module to achieve preliminary recognition of the target identity. The key pressure and trajectory consistency are calculated by the second determination module to determine whether there is an abnormal door opening mode, thereby further improving security. The judgment module comprehensively considers the abnormal mode and occlusion rate to confirm the target identity and authority. The identification module identifies the authority value according to the authority library and dynamically adjusts it. The correction module optimizes the judgment time and reduces misjudgment. The alarm module promptly issues an alarm based on the corrected data to ensure the intelligent protection and efficient response of the door lock, effectively solving the problem of low prevention effect in the face of complex security threats due to reliance on a single password input recognition.

[0047] Furthermore, by comparing and analyzing the real-time facial feature values ​​with the preset comparison feature values ​​through the feature comparison unit, the accuracy of target recognition can be effectively improved, ensuring that only targets that meet the preset facial feature standards will be identified as the first target, thereby reducing the risk of misidentification and improving the security and reliability of the system. Through real-time comparison and judgment, abnormal situations in the recognition process can be quickly responded to and handled.

[0048] Furthermore, by detecting abnormalities in key pressure and sliding trajectory, it can be determined whether there is abnormal door opening behavior, such as illegal use or unauthorized operation. By comparing with historical data, the system can more intelligently identify and respond to different abnormal situations, thereby improving the safety and reliability of door lock management.

[0049] Furthermore, by calculating the average and standard deviation of the historical key pressure within a preset historical time period, it is possible to effectively determine whether the real-time key pressure is consistent with the target's historical behavior, and then identify whether there is an abnormal key pattern. By comparing with the historical pressure data, the system can identify abnormal key behavior, thereby improving the security of the door lock management system and preventing unauthorized personnel from unlocking the door through abnormal operations. At the same time, the standard deviation of historical data can also help the system adapt to normal individual differences, ensuring that high-accuracy recognition can be maintained under a variety of operating habits.

[0050] Furthermore, by calculating the trajectory consistency, the system can accurately determine the similarity between the real-time sliding trajectory and the historical trajectory and identify whether there is abnormal behavior. Trajectory consistency calculation provides an efficient and accurate way to determine whether the user's operating habits are consistent with the historical behavior pattern, which helps to detect abnormal door opening modes, such as others imitating or forcibly opening the door. The advantage of this method is that it can effectively identify small trajectory deviations, improve the security and intelligence level of the door lock management system, and reduce the possibility of misjudgment.

[0051] Furthermore, by comparing the pressure consistency and trajectory consistency of real-time operation with the preset thresholds, abnormal door opening behavior can be accurately identified, avoiding the shortcomings of relying on a single judgment standard in traditional systems. The effectiveness of this process is reflected in the combination of multi-dimensional judgment standards (pressure and trajectory), which greatly improves the recognition accuracy and effectively avoids misidentification caused by misoperation or malicious intrusion. At the same time, the set threshold provides flexibility and can be dynamically adjusted according to the actual use environment, further enhancing the adaptability and security of the system.

[0052] Furthermore, by comprehensively analyzing the door opening mode and facial occlusion, abnormal behavior and occlusion effects can be effectively identified to avoid misidentification. By setting reasonable thresholds for abnormal number and occlusion rate, the system can ensure good fault tolerance for normal user operations, while improving sensitivity to security risks, which helps to enhance the security and intelligent recognition capabilities of door locks.

[0053] Furthermore, by monitoring the number of abnormal door opening patterns, frequent attempts to open the door abnormally can be identified in a timely manner, illegal or unauthorized entry can be prevented, and the security of the system can be enhanced.

[0054] Furthermore, by dynamically adjusting the comparison feature values ​​to cope with the differences between targets with different permissions, the system can more accurately identify targets in a variety of usage scenarios. The adjustment process can not only improve the accuracy of recognition, but also effectively cope with changes in user permissions or interference from the external environment on recognition, thereby enhancing the adaptability and security of the system.

[0055] Furthermore, by monitoring and correcting the fluctuation of the adjustment characteristic value, misjudgment caused by environmental or system fluctuations can be effectively avoided. By increasing the judgment time, a certain error can be tolerated, thereby improving the system's recognition accuracy of abnormal door opening patterns and enhancing the system's robustness and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 Schematic diagram of the smart door lock management system of this embodiment;

[0057] Figure 2 A decision logic diagram for determining a first target by a first determination unit for this implementation;

[0058] Figure 3 A decision logic diagram for determining an abnormal door opening mode by the second determination subunit of this implementation;

[0059] Figure 4 This is a decision logic diagram for the decision unit in this implementation to decide the second target. DETAILED DESCRIPTION

[0060] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0061] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0062] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0063] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0064] See also Figure 1 As shown, it is a schematic diagram of the smart door lock management system of this embodiment;

[0065] This embodiment provides an intelligent door lock management system, including:

[0066] A data acquisition module is used to collect the real-time key pressure and sliding trajectory of the target to be identified at the door lock during the door opening process, as well as the real-time image of the target to be identified;

[0067] An extraction module, used for extracting real-time facial feature values ​​and real-time facial occlusion rate of the real-time image;

[0068] A first determination module, connected to the extraction module, for determining a first target according to the real-time facial feature value and a preset comparison feature value;

[0069] a second determination module, which is connected to the data acquisition module and the first determination module respectively, and is used to determine an abnormal door opening mode according to the real-time key pressure and the real-time sliding trajectory of the first target;

[0070] a determination module, which is connected to the second determination module and the extraction module respectively, and is used to determine the second target according to the number of the abnormal door opening patterns and the real-time facial occlusion rate within a preset determination time;

[0071] an identification module, which is connected to the first determination module and the judgment module respectively, and is used to identify the authority value of the first target according to a preset authority library to form a first authority value, and to identify the authority value of the second target to form a second authority value;

[0072] an adjustment module, connected to the identification module and the first determination module respectively, for adjusting the preset comparison feature value according to the first authority value and the second authority value to form an adjusted feature value;

[0073] a correction module, which is connected to the adjustment module and the second determination module respectively, and is used to correct the preset determination time length according to the number of the adjustment characteristic values ​​within the preset correction time length;

[0074] An alarm module is connected to the determination module and is used to issue an intrusion alarm according to the second target determined based on the modified preset determination time.

[0075] The data acquisition module monitors the operation of the target to be identified during the door opening process in real time through sensors and cameras installed at the door lock. The key pressure is detected by the pressure sensor when the target presses the door lock button, and the sliding trajectory is recorded by the touch screen or trajectory sensing device. The real-time image captures the target's facial features through a high-definition camera to ensure that the image is clear and can provide accurate facial recognition data. The module can collect this information synchronously and provide data support for subsequent identity authentication and behavior analysis.

[0076] The extraction module uses a facial detection algorithm (such as Dlib) to identify and locate the face area in the real-time image, and then uses facial feature point extraction technology (such as Dlib's 68-point model) to obtain the key feature points of the face. By calculating the distance and geometric relationship between facial feature points, these features are converted into numerical facial feature values. At the same time, an occlusion detection algorithm is used to analyze whether the facial area is occluded, calculate the occlusion ratio, and obtain the numerical value of the real-time facial occlusion rate. Finally, the numerical data of facial feature values ​​and occlusion rates are used for subsequent target identity recognition and security determination.

[0077] The recognition module identifies the target's permission value based on the data in the preset permission library by connecting with the first determination module and the judgment module. The preset permission library contains different targets (such as personnel, equipment, etc.) and their corresponding permission level information, which usually includes personnel identity information, access rights, operation rights, etc. The recognition process matches the records in the preset permission library by comparing the facial features, keystroke behavior, and sliding trajectory of the target to be identified. When a match with a certain target feature is identified, the module will extract the corresponding permission value from the permission library, generate the permission level of the target, and use machine learning algorithms or image recognition technology to ensure accurate identification of the target's permissions through feature extraction and comparison.

[0078] The preset comparison feature value is a standard value used for comparison with the facial feature value collected in real time. It usually depends on the target facial feature data collected when the system is first registered. It is set to the average value or a specific range value of the target facial feature. In this embodiment, it is set to the standard value range of the target facial feature, which can ensure that the target identity can be accurately matched under normal circumstances.

[0079] The preset judgment time is a time window for judging abnormal door opening modes, which is usually determined by the time span of normal operation and is set between 1 second and 5 seconds. In this embodiment, it is set to 3 seconds, which can effectively identify abnormal behavior in a short period of time and will not miss abnormal modes due to slow reaction.

[0080] The preset correction time is the time period used to make corrections based on the fluctuations in the adjusted characteristic values. It usually depends on the system's tolerance for characteristic changes and is set between 2 seconds and 10 seconds. In this embodiment, it is set to 5 seconds, which can balance the accuracy and real-time nature of the correction and avoid excessive correction that may cause system instability.

[0081] The data acquisition module obtains the target's key pressure, sliding trajectory and facial image information in real time during the door opening process. The extraction module extracts facial feature values ​​and occlusion rates from the facial image, and the first determination module determines the target identity based on the facial feature values ​​and the preset comparison value. The second determination module combines the key pressure and sliding trajectory to determine whether there is an abnormal door opening mode. The judgment module further confirms the identity of the second target and identifies its authority based on the number of abnormal patterns and the facial occlusion rate. The adjustment module dynamically adjusts the comparison feature value according to the authority difference, and the correction module corrects the judgment time according to the historical data. Finally, the alarm module issues an intrusion alarm based on the corrected judgment result to ensure the safety of the door lock.

[0082] The security of the door lock is improved through multi-level identity authentication and abnormal behavior detection. The data acquisition module obtains key pressure, sliding trajectory and facial image in real time, and extracts facial features and occlusion rate in combination with the extraction module to achieve preliminary recognition of the target identity. The second determination module calculates the key pressure and trajectory consistency to determine whether there is an abnormal door opening mode, further improving security. The judgment module integrates abnormal patterns and occlusion rates to confirm the target identity and permissions. The recognition module identifies the permission value based on the permission library and makes dynamic adjustments. The correction module optimizes the judgment time to reduce misjudgment. The alarm module issues an alarm in time based on the corrected data to ensure the intelligent protection and efficient response of the door lock, effectively solving the problem of low prevention effect in the face of complex security threats due to reliance on single password input recognition.

[0083] Please continue reading Figure 2 As shown, it is a decision logic diagram of the first determination unit determining the first target in this implementation;

[0084] Specifically, the first determining module includes:

[0085] A feature comparison unit, used to compare the real-time facial feature value with the preset comparison feature value to form a feature comparison result;

[0086] A first determination unit is connected to the feature comparison unit, and is used to determine that the target to be identified is a first target when the feature comparison result is that the real-time facial feature value is greater than the preset comparison feature value.

[0087] The first determination module compares and analyzes the real-time facial feature value with the preset comparison feature value through the feature comparison unit. When the real-time facial feature value is greater than the preset comparison feature value, the first determination unit confirms that the target to be identified is the first target based on the feature comparison result. This process accurately determines the target identity by comparing the target facial features with the registered standard feature values.

[0088] By comparing and analyzing the real-time facial feature values ​​with the preset comparison feature values ​​through the feature comparison unit, the accuracy of target recognition can be effectively improved, ensuring that only targets that meet the preset facial feature standards will be identified as the first target, thereby reducing the risk of misidentification and improving the security and reliability of the system. Through real-time comparison and judgment, it can quickly respond to and handle abnormal situations in the recognition process.

[0089] Specifically, the second determination module includes:

[0090] A first calculation unit, configured to calculate a pressure consistency according to the real-time key pressure and the historical key pressure within a preset historical time period;

[0091] A second calculation unit is used to calculate the trajectory consistency according to the real-time sliding trajectory and the historical sliding trajectory within the preset historical time period;

[0092] A second determining unit is connected to the first calculating unit and the second calculating unit respectively, and is used to determine an abnormal door opening mode according to the pressure consistency and the trajectory consistency.

[0093] The preset historical duration is the time range of historical data used for reference when calculating the consistency of real-time key pressure and sliding trajectory. It is usually set to a longer period of time, such as one week to one month, so as to accurately reflect the regular behavior pattern of the target person. In this embodiment, the preset historical duration is set to 20 days to ensure that the system has enough historical data to judge the regular behavior of the target person and effectively distinguish abnormal behavior.

[0094] Historical key pressure refers to the recorded data of the key pressure applied by the target during the door opening process within a certain historical period. This data reflects the pressure value generated each time the target presses the door lock button, and its distribution characteristics can be statistically analyzed through pressure data at multiple time points. Usually, historical key pressure is calculated based on the normal behavior pattern of the target individual, and its typical pressing pressure range is established through statistical indicators such as the mean and standard deviation of historical data.

[0095] The historical sliding trajectory refers to the trajectory data formed by the sliding unlocking operation performed by the target during the door opening process within a certain historical length of time. This data records the target's finger trajectory when sliding the door lock screen, usually expressed in the form of continuous sliding points (i.e., the position coordinates at each time point). The historical sliding trajectory reflects the target's sliding gesture pattern, such as the shape, speed, direction, and other characteristics of the sliding path.

[0096] The second determination module calculates the pressure consistency between the real-time key pressure and the key pressure within the preset historical time length, and the trajectory consistency between the real-time sliding trajectory and the historical sliding trajectory through the first calculation unit and the second calculation unit. The pressure consistency and trajectory consistency calculation results will be input into the second determination unit to determine whether an abnormal door opening mode occurs according to the set threshold. When the consistency is lower than the preset standard, the system can identify abnormal door opening behavior.

[0097] By detecting abnormalities in key pressure and sliding trajectory, it can be determined whether there is abnormal door opening behavior, such as illegal use or unauthorized operation. By comparing with historical data, the system can more intelligently identify and respond to different abnormal situations, thereby improving the security and reliability of door lock management.

[0098] Specifically, the first computing unit includes:

[0099] A historical pressure calculation subunit, used to calculate an average value of historical key pressures within a preset historical time period to form an average historical pressure, and to calculate a standard deviation of historical key pressures within a preset historical time period to form a historical pressure standard deviation;

[0100] The pressure consistency calculation subunit is connected to the historical pressure calculation subunit, and is used to calculate the absolute value of the difference between the real-time key pressure and the average historical pressure, and to calculate the ratio of the absolute value to the historical pressure standard deviation to form pressure consistency.

[0101] The first calculation unit calculates the average value and standard deviation of the historical key pressure within the preset historical time length through the historical pressure calculation subunit, thereby forming the average historical pressure and the historical pressure standard deviation. Then, the pressure consistency calculation subunit compares the difference between the real-time key pressure and the average historical pressure, and calculates the ratio of the difference to the historical pressure standard deviation to obtain the pressure consistency value. This value is used to measure the consistency between the real-time key pressure and the historical normal pressure.

[0102] By calculating the average and standard deviation of the historical key pressure within the preset historical time period, it is possible to effectively determine whether the real-time key pressure is consistent with the target's historical behavior, and then identify whether there is an abnormal key pattern. By comparing with the historical pressure data, the system can identify abnormal key behavior, thereby improving the security of the door lock management system and preventing unauthorized personnel from unlocking the door through abnormal operations. At the same time, the standard deviation of historical data can also help the system adapt to normal individual differences, ensuring that high-accuracy recognition can be maintained under a variety of operating habits.

[0103] Specifically, the second computing unit includes:

[0104] The historical trajectory point calculation subunit is used to calculate the average value of the sliding trajectory points at each time point within a preset historical time length to form a number of average historical trajectory points;

[0105] A trajectory drawing subunit, connected to the historical trajectory point calculation subunit, for drawing an average historical sliding trajectory according to the average historical trajectory points;

[0106] The trajectory consistency calculation subunit is connected to the trajectory drawing subunit and is used to calculate the cosine similarity between the average historical sliding trajectory and the real-time sliding trajectory to form the trajectory consistency.

[0107] The second calculation unit calculates the average value of the sliding trajectory points at each time point within the preset historical duration through the historical trajectory point calculation subunit, thereby forming a number of average historical trajectory points. Then, the trajectory drawing subunit draws the average historical sliding trajectory based on these average historical trajectory points. Finally, the trajectory consistency calculation subunit calculates the cosine similarity between the real-time sliding trajectory and the average historical sliding trajectory to obtain a trajectory consistency value, which measures the consistency between the real-time trajectory and the historical trajectory.

[0108] By calculating the trajectory consistency, the system can accurately determine the similarity between the real-time sliding trajectory and the historical trajectory and identify whether there is abnormal behavior. Trajectory consistency calculation provides an efficient and accurate way to determine whether the user's operating habits are consistent with the historical behavior pattern, which helps to detect abnormal door opening modes, such as others imitating or forcibly opening the door. The advantage of this method is that it can effectively identify small trajectory deviations, improve the security and intelligence level of the door lock management system, and reduce the possibility of misjudgment.

[0109] Please continue reading Figure 3 As shown, it is a decision logic diagram of the second determination subunit determining the abnormal door opening mode in this implementation;

[0110] Specifically, the second determining unit includes:

[0111] A pressure comparison subunit, used to compare the pressure consistency with a preset pressure consistency threshold to form a pressure comparison result;

[0112] A trajectory comparison subunit, used to compare the trajectory consistency with a preset trajectory consistency threshold to form a trajectory comparison result;

[0113] The second determination subunit is connected to the pressure comparison subunit and the trajectory comparison subunit respectively, and is used to determine that an abnormal door opening mode occurs when the pressure comparison result is that the pressure consistency is less than a preset pressure consistency threshold, and the trajectory comparison result is that the trajectory consistency is less than a preset trajectory consistency threshold.

[0114] The preset pressure consistency threshold is a standard for judging the difference between real-time key pressure and historical key pressure. It depends on the pressure range applied by the key when the user opens the door normally and the sensitivity of the device. It is usually set between 0.2 and 0.5. In this embodiment, it is set to 0.3, which can effectively balance the gap between normal user operations and abnormal behaviors, avoid misjudging normal operations as abnormalities, and has moderate sensitivity to avoid excessive screening.

[0115] The preset trajectory consistency threshold is used to judge the similarity between the real-time sliding trajectory and the historical sliding trajectory. It depends on the normal sliding trajectory characteristics of the user and the trajectory recognition accuracy of the device. It is usually set between 0.7 and 0.9. In this embodiment, it is set to 0.8, which ensures high-precision recognition of normal and abnormal trajectories and avoids misidentification. At the same time, it has good adaptability to slight trajectory differences between different users.

[0116] The second determination unit compares the calculated pressure consistency with the preset pressure consistency threshold through the pressure comparison subunit to form a pressure comparison result; at the same time, the trajectory comparison subunit compares the calculated trajectory consistency with the preset trajectory consistency threshold to form a trajectory comparison result. The second determination subunit determines that an abnormal door opening mode exists based on the pressure comparison result and the trajectory comparison result when the pressure consistency is lower than the preset pressure consistency threshold and the trajectory consistency is lower than the preset trajectory consistency threshold.

[0117] By comparing the pressure consistency and trajectory consistency of real-time operation with the preset threshold, abnormal door opening behavior can be accurately identified, avoiding the shortcomings of relying on a single judgment standard in traditional systems. The effectiveness of this process is reflected in the combination of multi-dimensional judgment standards (pressure and trajectory), which greatly improves the recognition accuracy and effectively avoids misidentification caused by misoperation or malicious intrusion. At the same time, the set threshold provides flexibility and can be dynamically adjusted according to the actual use environment, further enhancing the adaptability and security of the system.

[0118] Please continue reading Figure 4 As shown, it is a decision logic diagram of the decision unit for deciding the second target in this implementation;

[0119] Specifically, the determination module includes:

[0120] An abnormality comparison unit, used to compare the number of abnormal door opening modes with a preset abnormal number threshold to form an abnormality comparison result;

[0121] An occlusion comparison unit, used to compare the real-time facial occlusion rate with a preset occlusion rate threshold to form an occlusion comparison result;

[0122] A determination unit is connected to the abnormal comparison unit and the occlusion comparison unit, respectively, and is used to determine that the first target is the second target when the abnormal comparison result is that the number of the abnormal door opening patterns is less than the preset abnormal number threshold, and the real-time facial occlusion rate is greater than the preset occlusion rate threshold.

[0123] The preset occlusion rate threshold is used to determine the degree of occlusion in the real-time facial image, and is usually set according to the accuracy of the facial recognition algorithm and the tolerance of the device. This value is usually set between 30% and 50%, and is set to 40% in this embodiment. By setting a moderate occlusion rate threshold, the tolerance for slight occlusion and the recognition of important occlusion can be effectively balanced to avoid misjudgment.

[0124] The preset abnormal number threshold is used to determine the number of abnormal door opening modes, which is usually set according to the security requirements of the door lock system and the user's behavior pattern, and is generally set between 1 and 3 times. In this embodiment, it is set to 2 times. By setting a reasonable abnormal number threshold, it is possible to avoid interference with occasional misoperations, while enhancing sensitivity to potential security risks and reducing the occurrence of false alarms.

[0125] The determination module determines whether there is abnormal door opening behavior by comparing the number of abnormal door opening patterns with the preset abnormal number threshold; at the same time, it compares the real-time facial occlusion rate with the preset occlusion rate threshold to evaluate whether the facial features are blocked. If the number of abnormal door opening patterns is less than the preset threshold, and the real-time facial occlusion rate is greater than the preset threshold, the system determines that the first target is the second target and identifies potential identity or security risks.

[0126] By comprehensively analyzing the door opening mode and facial occlusion, it can effectively identify abnormal behaviors and occlusion effects to avoid misidentification. By setting reasonable thresholds for abnormal number and occlusion rate, it ensures that the system has good fault tolerance for normal user operations, while improving sensitivity to security risks, which helps to enhance the security and intelligent recognition capabilities of door locks.

[0127] Specifically, the determination unit is further configured to determine that the target to be identified has no access permission when the abnormal comparison result is that the number of the abnormal door opening patterns is greater than the preset abnormal number threshold.

[0128] During the operation of the judgment unit, when the number of abnormal door opening patterns is greater than the preset abnormal number threshold, it indicates that there is abnormal behavior or potential security threat. At this time, the judgment unit will determine that the target to be identified does not have access permission based on this result, and issue a corresponding alarm or refuse to open the door. By comparing with the abnormal number threshold, the system can effectively identify situations where abnormal door opening behaviors occur frequently, thereby improving the security of the door lock system.

[0129] By monitoring the number of abnormal door opening patterns, frequent attempts to open the door abnormally can be identified in a timely manner, illegal or unauthorized entry can be prevented, and the security of the system can be enhanced.

[0130] Specifically, the adjustment module includes:

[0131] a ratio calculation unit, used to calculate a relative deviation between the first authority value and the second authority value to form an adjustment ratio;

[0132] a ratio comparison unit, connected to the ratio calculation unit, for comparing the adjustment ratio with a preset ratio threshold to form a ratio comparison result;

[0133] An adjustment unit is connected to the ratio comparison unit and is used to increase the preset comparison characteristic value according to the adjustment ratio and the preset adjustment coefficient to form an adjustment characteristic value when the ratio comparison result is that the adjustment ratio is greater than the preset ratio threshold, and the adjustment ratio and the adjustment characteristic value are positively correlated.

[0134] The preset ratio threshold is used to determine the relative deviation between the first permission value and the second permission value, which depends on the precise requirements of permission identification and the fault tolerance of the system. It is usually set between 0.05 and 0.2, and is set to 0.1 in this embodiment. The advantage of this setting is that it can avoid triggering unnecessary adjustments due to small deviations while tolerating a certain error, thereby ensuring the stability of system identification.

[0135] The preset adjustment coefficient is used to adjust the multiple of the comparison feature value, which depends on the adjustment range and recognition accuracy required by the system, and is usually set between 1 and 3. In this embodiment, it is set to 1.5. The advantage of such a setting is that it can improve the flexibility and accuracy of the recognition system to a certain extent, avoid excessive deviation of the feature value caused by excessive adjustment, and thus maintain the accuracy of recognition.

[0136] First, the relative deviation between the first authority value and the second authority value is calculated by the ratio calculation unit to obtain the adjustment ratio. Then, the ratio comparison unit compares the adjustment ratio with the preset ratio threshold to determine whether it exceeds the preset threshold. If the adjustment ratio is greater than the threshold, the adjustment unit adjusts the preset comparison feature value according to the ratio and the preset adjustment coefficient to form a new adjustment feature value. Through this process, the system can dynamically adjust the comparison feature value according to the difference in the target's authority value, thereby improving recognition accuracy and flexibility.

[0137] By dynamically adjusting the comparison feature values ​​to cope with the differences between targets with different permissions, the system can more accurately identify targets in a variety of usage scenarios. The adjustment process can not only improve the accuracy of recognition, but also effectively cope with changes in user permissions or interference from the external environment on recognition, thereby enhancing the adaptability and security of the system.

[0138] Specifically, the correction module includes:

[0139] A fluctuation calculation unit, used to calculate the standard deviation of the quantity of the adjusted characteristic value to form a quantity fluctuation value;

[0140] A correction unit is connected to the fluctuation calculation unit and is used to increase the preset judgment time according to the relative deviation between the quantity fluctuation value and the preset fluctuation value threshold and a preset correction coefficient when the quantity fluctuation value is greater than the preset fluctuation value threshold, and the relative deviation and the preset judgment time are positively correlated.

[0141] The correction module calculates the standard deviation of the number of adjusted characteristic values ​​through the fluctuation calculation unit to obtain the quantity fluctuation value. When the quantity fluctuation value is greater than the preset fluctuation value threshold, the correction unit increases the preset judgment time according to the relative deviation between the quantity fluctuation value and the preset fluctuation value threshold, combined with the preset correction coefficient. This process is designed to dynamically adjust the judgment time to adapt to the uncertainty in the recognition environment and ensure the accuracy of the judgment result.

[0142] By monitoring and correcting the fluctuation of the adjusted characteristic value, misjudgment caused by environmental or system fluctuations can be effectively avoided. By increasing the judgment time, a certain error can be tolerated, thereby improving the system's recognition accuracy of abnormal door opening patterns and enhancing the system's robustness and adaptability.

[0143] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0144] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent door lock management system, characterized in that: include: A data acquisition module is used to collect the real-time key pressure and sliding trajectory of the target to be identified at the door lock during the door opening process, as well as the real-time image of the target to be identified; An extraction module, used for extracting real-time facial feature values ​​and real-time facial occlusion rate of the real-time image; A first determination module, connected to the extraction module, for determining a first target according to the real-time facial feature value and a preset comparison feature value; a second determination module, which is connected to the data acquisition module and the first determination module respectively, and is used to determine an abnormal door opening mode according to the real-time key pressure and the real-time sliding trajectory of the first target; a determination module, which is connected to the second determination module and the extraction module respectively, and is used to determine the second target according to the number of the abnormal door opening patterns and the real-time face occlusion rate within a preset determination time; an identification module, which is connected to the first determination module and the judgment module respectively, and is used to identify the authority value of the first target according to a preset authority library to form a first authority value, and to identify the authority value of the second target to form a second authority value; an adjustment module, connected to the identification module and the first determination module respectively, for adjusting the preset comparison feature value according to the first authority value and the second authority value to form an adjusted feature value; a correction module, which is connected to the adjustment module and the second determination module respectively, and is used to correct the preset determination time length according to the number of the adjustment characteristic values ​​within the preset correction time length; An alarm module is connected to the determination module and is used to issue an intrusion alarm according to the second target determined based on the modified preset determination time.

2. The intelligent door lock management system according to claim 1, characterized in that: The first determining module comprises: A feature comparison unit, used to compare the real-time facial feature value with the preset comparison feature value to form a feature comparison result; A first determination unit is connected to the feature comparison unit, and is used to determine that the target to be identified is a first target when the feature comparison result is that the real-time facial feature value is greater than the preset comparison feature value.

3. The intelligent door lock management system according to claim 2, characterized in that: The second determining module comprises: A first calculation unit, configured to calculate a pressure consistency according to the real-time key pressure and the historical key pressure within a preset historical time period; A second calculation unit is used to calculate the trajectory consistency according to the real-time sliding trajectory and the historical sliding trajectory within the preset historical time period; A second determining unit is connected to the first calculating unit and the second calculating unit respectively, and is used to determine an abnormal door opening mode according to the pressure consistency and the trajectory consistency.

4. The intelligent door lock management system according to claim 3, characterized in that: The first computing unit comprises: A historical pressure calculation subunit, used to calculate an average value of historical key pressures within a preset historical time period to form an average historical pressure, and to calculate a standard deviation of historical key pressures within a preset historical time period to form a historical pressure standard deviation; The pressure consistency calculation subunit is connected to the historical pressure calculation subunit, and is used to calculate the absolute value of the difference between the real-time key pressure and the average historical pressure, and to calculate the ratio of the absolute value to the historical pressure standard deviation to form pressure consistency.

5. The intelligent door lock management system according to claim 4, characterized in that: The second computing unit comprises: The historical trajectory point calculation subunit is used to calculate the average value of the sliding trajectory points at each time point within a preset historical time length to form a number of average historical trajectory points; A trajectory drawing subunit, connected to the historical trajectory point calculation subunit, for drawing an average historical sliding trajectory according to the average historical trajectory points; The trajectory consistency calculation subunit is connected to the trajectory drawing subunit and is used to calculate the cosine similarity between the average historical sliding trajectory and the real-time sliding trajectory to form the trajectory consistency.

6. The intelligent door lock management system according to claim 5, characterized in that: The second determining unit includes: A pressure comparison subunit, used to compare the pressure consistency with a preset pressure consistency threshold to form a pressure comparison result; A trajectory comparison subunit, used to compare the trajectory consistency with a preset trajectory consistency threshold to form a trajectory comparison result; The second determination subunit is connected to the pressure comparison subunit and the trajectory comparison subunit respectively, and is used to determine that an abnormal door opening mode occurs when the pressure comparison result is that the pressure consistency is less than a preset pressure consistency threshold, and the trajectory comparison result is that the trajectory consistency is less than a preset trajectory consistency threshold.

7. The intelligent door lock management system according to claim 6, characterized in that: The determination module comprises: An abnormality comparison unit, used to compare the number of abnormal door opening modes with a preset abnormal number threshold to form an abnormality comparison result; An occlusion comparison unit, used to compare the real-time facial occlusion rate with a preset occlusion rate threshold to form an occlusion comparison result; A determination unit is connected to the abnormal comparison unit and the occlusion comparison unit, respectively, and is used to determine that the first target is the second target when the abnormal comparison result is that the number of the abnormal door opening patterns is less than the preset abnormal number threshold, and the real-time facial occlusion rate is greater than the preset occlusion rate threshold.

8. The intelligent door lock management system according to claim 7, characterized in that: The determination unit is further configured to determine that the target to be identified has no access permission when the abnormal comparison result is that the number of the abnormal door opening patterns is greater than the preset abnormal number threshold.

9. The intelligent door lock management system according to claim 8, characterized in that: The adjustment module comprises: a ratio calculation unit, used to calculate a relative deviation between the first authority value and the second authority value to form an adjustment ratio; a ratio comparison unit, connected to the ratio calculation unit, for comparing the adjustment ratio with a preset ratio threshold to form a ratio comparison result; The adjustment unit is connected to the ratio comparison unit, and is used to increase the preset comparison characteristic value according to the adjustment ratio and a preset adjustment coefficient to form an adjustment characteristic value when the ratio comparison result is that the adjustment ratio is greater than the preset ratio threshold.

10. The intelligent door lock management system according to claim 9, characterized in that: The correction module comprises: A fluctuation calculation unit, used to calculate the standard deviation of the quantity of the adjusted characteristic value to form a quantity fluctuation value; A correction unit is connected to the fluctuation calculation unit and is used to increase the preset determination time according to a relative deviation between the quantity fluctuation value and the preset fluctuation value threshold and a preset correction coefficient when the quantity fluctuation value is greater than the preset fluctuation value threshold.

Images