Multi-mode switching intelligent door lock safety control method and device

Abstract

The invention relates to a multi-mode switching intelligent door lock safety control method and device, and the method comprises the following steps: collecting the identity information of a user, obtaining the biological feature data of the user, and carrying out the feature coding of the biological feature data of the user, and obtaining a biological feature coding sequence; performing similarity matching calculation on the biological feature coding sequence and a pre-stored user identity database to obtain an identity matching result; performing verification mode selection on the door lock based on the identity matching result to generate a current verification mode; and state control is conducted on the door lock based on the current verification mode, door lock state control operation is completed, and the technical problem that an existing intelligent door lock mostly adopts a single verification mode, and the recognition rate is low when the existing intelligent door lock faces different user groups and complex use scenes is solved.

Description

Technical Field

[0001] This invention relates to the field of door lock technology, and in particular to a smart door lock security control method and device with multi-mode switching. Background Technology

[0002] With the rapid development of smart home technology, smart door locks, as the first line of defense for home security, are receiving increasing attention for their security and convenience. Traditional mechanical locks, due to issues such as easy loss and duplication of keys, can no longer meet the security needs of modern families. Therefore, smart door lock systems based on biometric recognition are gradually becoming mainstream, using fingerprints, faces, iris scans, and other methods for identity verification, improving the security level and ease of use of door locks. However, most existing smart door locks adopt a single verification mode, which suffers from low recognition rates, slow response speeds, and insufficient security when facing different user groups and complex usage scenarios. Summary of the Invention

[0003] The purpose of this invention is to at least partially solve one of the technical problems existing in the prior art.

[0004] To achieve the above objectives, in a first aspect, the present invention provides a multi-mode switching smart door lock security control method, comprising the following steps: Collect user biometric data and perform feature encoding on the user biometric data to obtain a biometric encoding sequence; The biometric encoding sequence is matched with a pre-stored user identity database to obtain the identity matching result; Based on the identity matching result, the door lock is selected for verification mode, and the current verification mode is generated. Based on the current verification mode, the door lock is state controlled to complete the door lock state control operation.

[0005] Furthermore, the process of collecting user biometric data and performing feature encoding on the user biometric data to obtain a biometric encoding sequence includes: The user's original biometric data is collected by an identity recognition module installed on the door lock, and noise suppression processing is performed on the original biometric data to obtain a standardized biometric image, which is then used as the user's biometric data. The distribution of local feature points is extracted from the user's biometric data, feature point coordinates are generated, and geometric invariant transformation and normalization are performed on the feature point coordinates to generate a feature description vector at a uniform scale. Binarization encoding and bit string compression operations are performed on each vector in the feature description vector to obtain the biometric feature encoding sequence.

[0006] Furthermore, the step of matching the biometric encoded sequence with a pre-stored user identity database to obtain an identity matching result includes: Obtain the feature index table corresponding to the standard biometric code that is pre-stored in the pre-stored user identity database, and convert the format of the biometric code sequence to obtain the code string to be matched; Based on the feature index table, the string to be matched is initially screened to obtain candidate matching codes, and bitwise operations are performed on the candidate matching codes and the string to be matched to generate a bitwise comparison result. The user's identity matching result is determined by comparing the encoded bits.

[0007] Furthermore, the preliminary screening of the code string to be matched based on the feature index table to obtain candidate matching codes includes: The hash value of the encoding string to be matched is extracted to obtain the hash value to be matched, and the standard biometric codes in the feature index table are mapped by hash index block to obtain the hash index table; Based on the hash value to be matched, an approximate value lookup is performed on the hash index table to obtain a set of approximate hash values, and the approximate hash value set is encoded and traced to obtain a traceability code; The source code is partitioned for matching to obtain the partition matching result, and the source code is initially filtered based on the partition matching result to obtain the filtered code; The feature bit weights of the filtering codes are calculated to obtain the coding weight values, and the filtering codes are sorted and filtered based on the coding weight values ​​to obtain candidate matching codes.

[0008] Furthermore, the step of performing bitwise operations to compare the candidate matching code with the code string to be matched, and generating a code difference value, includes: Each coding sequence in the candidate matching code is parsed by bit segment to obtain a hierarchical coding structure, and the hierarchical coding structure is position-calibrated to obtain a calibrated coding array; Based on the calibrated encoding array, a bit-by-bit XOR comparison is performed on the encoding string to be matched to obtain a bit difference comparison matrix, and the difference bit statistics are performed on the bit difference comparison matrix to obtain the difference bit distribution characteristics. Hamming distance quantization is performed on the differential bit distribution features to obtain distance metric parameters, and differential values ​​are calculated based on the distance metric parameters to obtain encoded differential values.

[0009] Furthermore, the step of selecting a verification mode for the door lock based on the identity matching result and generating the current verification mode includes: Based on the identity matching results, the user's security level is assessed to obtain the user security risk coefficient, and based on the user security risk coefficient, the permission level is calculated to obtain the user permission level parameter. Based on the user permission level parameters, a query is performed in the door lock's verification strategy library to obtain candidate verification modes, and the candidate verification modes are sorted by priority to obtain the current verification mode.

[0010] Furthermore, the step of controlling the door lock state based on the current verification mode to complete the door lock state control operation includes: The current verification mode is mapped to control permissions to obtain a door lock operation permission code, and the door lock operation permission code is verified for security level to obtain an authorized execution signal; Based on the authorized execution signal, the lock cylinder drive mechanism of the door lock is controlled to obtain the lock cylinder rotation control sequence; Based on the lock cylinder rotation control sequence, the torque of the lock cylinder drive mechanism is adjusted to obtain the lock cylinder action execution parameters; The mechanical state data of the door lock is analyzed by the lock cylinder action execution parameters, and the opening and closing status is identified based on the mechanical state data to obtain the current status identifier of the door lock; The control result is confirmed based on the current status identifier of the door lock, a status control signal is obtained, and the door lock status control operation is completed based on the status control signal.

[0011] Secondly, a multi-mode switching smart door lock security control device includes: The acquisition module is used to collect the user's biometric data and perform feature encoding on the user's biometric data to obtain a biometric encoding sequence; The calculation module is used to perform matching calculations between the biometric encoding sequence and the pre-stored user identity database to obtain the identity matching result; The selection module is used to select the verification mode of the door lock based on the identity matching result and generate the current verification mode. The control module is used to control the state of the door lock based on the current verification mode and complete the door lock state control operation.

[0012] Thirdly, embodiments of the present invention provide a multi-mode switching smart door lock security control device, comprising: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the multi-mode switching smart door lock security control method.

[0013] Fourthly, embodiments of the present invention provide a storage medium storing a processor-executable program, which, when executed by a processor, is used to implement the aforementioned multi-mode switching smart door lock security control method.

[0014] This invention provides a multi-mode switching smart lock security control method, comprising the following steps: collecting user identity information to obtain user biometric data, and performing feature encoding on the user biometric data to obtain a biometric encoding sequence; performing matching calculations between the biometric encoding sequence and a pre-stored user identity database to obtain an identity matching result; selecting a verification mode for the lock based on the identity matching result to generate a current verification mode; and performing state control on the lock based on the current verification mode to complete the lock state control operation. This method solves the problems of low recognition rate, slow response speed, and insufficient security in existing smart locks that mostly adopt a single verification mode when facing different user groups and complex usage scenarios. It achieves a technical effect of balancing security and user experience in different scenarios by using a low-strength verification method to improve convenience when the identity matching degree is high, and automatically switching to a high-strength verification method to ensure security when the matching degree is low. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the steps of a smart door lock security control method with multi-mode switching in one embodiment of the present invention; Figure 2 This is a structural block diagram of a multi-mode switching smart door lock security control device according to an embodiment of the present invention; Figure 3 This is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0016] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] The embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention. The step numbers in the following embodiments are set only for ease of explanation, and there is no limitation on the order between the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0018] The following describes in detail, with reference to the accompanying drawings, a multi-mode switching smart door lock security control method and device according to an embodiment of the present invention. First, the multi-mode switching smart door lock security control method according to an embodiment of the present invention will be described with reference to the accompanying drawings.

[0019] like Figure 1 As shown, Figure 1 This invention provides a multi-mode switching smart door lock security control method, comprising the following steps: Step S1: Collect user biometric data and perform feature encoding on the user biometric data to obtain a biometric encoding sequence; Step S2: Match the biometric encoding sequence with the pre-stored user identity database to obtain the identity matching result; Step S3: Select a verification mode for the door lock based on the identity matching result and generate the current verification mode; Step S4: Perform state control on the door lock based on the current verification mode to complete the door lock state control operation.

[0020] Specifically, in the implementation of the smart door lock system, this solution completes user identity verification and door lock status control through a series of structured steps, including four key steps: collecting user biometric data and encoding the biometric data to obtain a biometric encoding sequence; matching the biometric encoding sequence with a pre-stored user identity database to obtain an identity matching result; selecting a verification mode for the door lock based on the identity matching result to generate the current verification mode; and controlling the door lock status based on the current verification mode to complete the door lock status control operation. The entire process is based on user biometrics, combined with database comparison and dynamic verification strategies to achieve intelligent control and enhanced security level of the door lock. In step S1, the system first collects user biometric data, which typically includes fingerprints, face, iris, and other biometric information. The collection devices can be fingerprint recognition modules, cameras, infrared sensors, etc., which record the user's current biometrics in the form of images, point clouds, or waveforms. After collection, the system needs to perform feature extraction and encoding processing on these raw data to generate a unique and identifiable biometric encoding sequence. For example, in fingerprint recognition, the system extracts key features such as fingerprint ridges and bifurcation points, and converts them into digital codes. In face recognition, it may extract geometric features such as facial contours, eye spacing, and nose shape, and generate feature vectors through deep learning models. The purpose of this process is to compress complex biometric data into structured digital codes, facilitating subsequent matching and comparison operations. In step S2, the system performs matching calculations between the biometric code sequence generated in step S1 and a pre-stored user identity database to obtain the identity matching result. The user identity database stores the biometric codes of registered users and their corresponding user identity information, such as user ID, permission level, and family member relationships. The matching process typically uses similarity algorithms or classification algorithms, such as Euclidean distance calculation, cosine similarity, support vector machine (SVM), or neural network models. By calculating the similarity between the current biometric code and existing codes in the database, it determines whether they belong to the same user. If the similarity is higher than a preset threshold, the match is considered successful; otherwise, the match is considered unsuccessful. For example, when a user attempts to unlock the door using their fingerprint, the system extracts their fingerprint features and compares them with multiple fingerprint codes in the database. If a record with a similarity exceeding a set value is found, the system returns the user's ID as the identity matching result. In step S3, based on the identity matching result obtained in step S2, the system selects a verification mode for the door lock and generates the current verification mode. This verification mode is dynamically determined based on the user's identity information, and different users may correspond to different verification methods.For example, for family member A, the system may only require fingerprint verification to unlock the door; while for visitor B, the system may require both facial recognition and mobile verification code verification to enhance security. The core of this dynamic verification mechanism lies in flexibly configuring the verification method based on factors such as user identity, usage scenario, and time period. For instance, during peak daytime hours when family members frequently enter and exit, the system can choose a more convenient verification mode, such as single fingerprint verification; while at night or when the system detects abnormal behavior, it may switch to a multi-verification mode, such as fingerprint + facial recognition + dynamic password, to enhance security. This mechanism not only improves the system's adaptability but also balances security and convenience. In step S4, the system performs state control on the door lock based on the current verification mode generated in step S3, completing the door lock state control operation. State control includes various operations such as unlocking, locking, alarm, and temporary authorization. After the user completes all verification steps in the current verification mode, the system will send an unlocking command to the door lock actuator, driving the motor or electromagnetic lock to complete the unlocking action; if verification fails, the system may trigger an alarm mechanism or record abnormal attempts and push relevant information to the user's mobile phone via the network module. For example, when a family member uses their fingerprint to unlock the door, the system recognizes that the fingerprint belongs to a registered user and the current verification mode is single fingerprint verification. Upon successful verification, the system immediately executes the unlocking operation. However, if an unregistered user attempts to unlock the door, the system may automatically switch to temporary authorization mode after verification failure, allowing the user to unlock the door by entering a temporary password or waiting for remote authorization from a family member. To further illustrate the practical application of this solution, consider a typical family scenario: A user returns home from get off work and approaches the door lock. The system detects the user's approach using an infrared sensor and activates the camera for facial recognition. After acquiring the user's facial image, the system extracts facial features and generates a biometric encoding sequence. This sequence is then matched against data in the user identity database, identifying the user as a family member whose verification mode is "fingerprint + face" dual verification. Therefore, the system enters this verification mode and prompts the user to register their fingerprint. After the user places their finger in the fingerprint recognition area, the system again collects the fingerprint features, encodes them, and compares them with data in the database. Once confirmed, the system determines that the identity match is successful and executes the unlocking operation, automatically opening the door. The entire process was smooth and efficient, ensuring both security and convenience. In another scenario, a visitor arrives at the door and attempts to call the homeowner via the doorbell. The homeowner remotely verifies the visitor's identity via a mobile app, generates a temporary authorization code, and sends it to the visitor. The visitor enters the temporary authorization code on the door lock keypad, and the system recognizes it as a temporary authorization, subsequently initiating a verification process involving a temporary password and facial recognition.The system collects the visitor's facial information, encodes its features, and compares them with temporary visitor templates in the database. If they match, the system determines that the verification is successful and executes the unlocking operation. This scenario demonstrates the system's flexibility and security under different user identities and usage scenarios, dynamically adjusting the verification method according to the needs of different users, thereby improving the overall user experience.

[0021] In a specific scenario, the process of collecting user biometric data and encoding the user biometric data to obtain a biometric encoding sequence includes: The user's original biometric data is collected by an identity recognition module installed on the door lock, and noise suppression processing is performed on the original biometric data to obtain a standardized biometric image, which is then used as the user's biometric data. The distribution of local feature points is extracted from the user's biometric data, feature point coordinates are generated, and geometric invariant transformation and normalization are performed on the feature point coordinates to generate a feature description vector at a uniform scale. Binarization encoding and bit string compression operations are performed on each vector in the feature description vector to obtain the biometric feature encoding sequence.

[0022] Specifically, the identity recognition module, as the core hardware for collecting users' raw biometric data, typically includes devices such as fingerprint sensors, infrared cameras, and 3D structured light modules. These devices collect corresponding raw data based on different biometric types. For example, in fingerprint recognition, the fingerprint sensor collects images of the ridge and valley distribution of the user's fingerprint; in face recognition, the camera captures two-dimensional or three-dimensional image information of the user's face. The collected raw biometric data usually contains noise, such as uneven finger dryness or dirt interference in fingerprint images, or image blurring caused by uneven lighting in face recognition. Therefore, the system performs noise suppression processing on the raw biometric data to remove interfering factors in the image and enhance the clarity of key features, thereby obtaining a standardized biometric image as input data for subsequent processing. After obtaining the standardized biometric image, the system further extracts the distribution of local feature points in the image. Taking fingerprint recognition as an example, the system detects key feature points such as ridge endpoints and bifurcation points in the fingerprint image and generates corresponding feature point coordinates; while in face recognition, the system may extract the position coordinates of key facial points such as the corners of the eyes, the tip of the nose, and the corners of the mouth. These feature point coordinates constitute a local description of a user's biometrics. However, due to differences in angle, distance, or posture during the data collection process, these coordinates typically lack a uniform scale and orientation. Therefore, the system needs to perform geometrically invariant transformations (such as rotation, translation, and scaling invariance processing) and normalization on the feature point coordinates. This ensures that feature points acquired under different collection conditions can be mapped to a unified coordinate space, thereby generating a feature description vector at a uniform scale. This vector can more accurately represent the user's biometrics while reducing the impact of environmental and posture changes on the recognition results. After generating the feature description vector at a uniform scale, the system further encodes these vectors to facilitate subsequent database matching and rapid retrieval. Specifically, the system performs binarization encoding on each vector in the feature description vector, converting continuous feature values ​​into binary bit strings composed of 0s and 1s. This binarization process not only reduces data storage space but also improves the efficiency of matching calculations. Subsequently, the system performs bit string compression on these binary bit strings to remove redundant information and generate a compact biometric encoding sequence. This sequence will serve as a digital identifier of the user's identity and will be used for subsequent identity matching and verification operations. For example, in a home use scenario, when a user approaches the door lock to unlock it, the system activates the fingerprint recognition module to acquire their fingerprint image. The acquired raw fingerprint image may contain smudges and blurriness; the system performs noise suppression processing to obtain a clear, standardized fingerprint image. Subsequently, the system extracts fingerprint feature points from the image, such as endpoints and bifurcation points, and generates corresponding coordinate information. To eliminate the influence of differences in the user's finger placement angle and pressure, the system performs geometrically invariant transformations and normalization on these coordinates to generate feature description vectors at a uniform scale.Next, the system converts these vectors into binary bit strings and compresses them to generate the user's biometric encoding sequence. This sequence is then compared with existing data in the user identity database to complete identity verification.

[0023] In a specific scenario, the step of matching the biometric encoding sequence with a pre-stored user identity database to obtain an identity matching result includes: Obtain the feature index table corresponding to the standard biometric code that is pre-stored in the pre-stored user identity database, and convert the format of the biometric code sequence to obtain the code string to be matched; Based on the feature index table, the string to be matched is initially screened to obtain candidate matching codes, and bitwise operations are performed on the candidate matching codes and the string to be matched to generate a bitwise comparison result. The user's identity matching result is determined by comparing the encoded bits.

[0024] Specifically, to implement this matching process, the system first needs to access a pre-stored user identity database. This database stores standard biometric codes for multiple registered users. These codes are generated during the user registration phase by collecting data through the same identity recognition module and undergoing feature encoding processing, ensuring uniqueness and identifiability. Simultaneously, the database maintains a feature index table corresponding to these standard biometric codes. This table records key feature information for each biometric code in an indexed structure for rapid retrieval during the matching process. The feature index table is typically designed based on techniques such as hash indexes, inverted indexes, or near-nearest neighbor indexes, enabling efficient initial screening even with large-scale user data. After obtaining the feature index table, the system converts the format of the currently collected and processed biometric code sequences to generate a matching string. This format conversion ensures that the matching string matches the storage format of the standard biometric codes in the database, facilitating subsequent comparison operations. For example, if the standard biometric codes are stored as fixed-length binary bit strings, the matching string must also be converted to the same length binary format to ensure accurate comparison at the bitwise operation level. Next, the system performs preliminary screening of the code string to be matched based on the feature index table, identifying candidate matching codes that may be matched. This process is achieved through feature label or key bit matching in the feature index table. For example, if the feature values ​​of certain key bits in the code string to be matched are similar to the index features of a standard biometric code, then that standard code will be included in the candidate matching code set. Although this screening stage does not have the ability to finally confirm identity, it can significantly narrow down the scope of subsequent precise comparisons, thereby improving the overall matching efficiency. Subsequently, the system performs bitwise operations to compare the candidate matching codes with the code string to be matched, generating a bitwise comparison result. Bitwise comparison typically uses XOR operations or Hamming distance calculations, judging their similarity by comparing the number of bits that differ between the two binary bit strings. For example, if the two code strings are the same in most bits and have only a few differences, it means they may belong to the same user; conversely, if there are many differences, it means the match has failed. The system will judge the bitwise comparison result according to a preset similarity threshold. If the similarity is higher than the threshold, the match is considered successful; otherwise, it is considered a failure. Ultimately, the system uses the bit comparison results to determine the user's identity. If the match is successful, the system will return the corresponding user identity information, including user ID, permission level, etc., as the basis for subsequent door lock verification mode selection; if the match fails, the system may prompt the user to re-collect biometric features, or switch to an alternative verification method (such as temporary password, remote authorization, etc.) to ensure that legitimate users are not denied access due to identification errors.For example, in a typical home scenario, when a user approaches the door lock, the system activates the facial recognition module to capture their facial image and performs steps such as noise suppression, feature extraction, and encoding compression to generate a biometric encoding sequence. Subsequently, the system accesses the feature index table in the user's identity database and converts the encoding sequence into a standard format matching string. Through preliminary screening using the feature index table, the system extracts several candidate matching codes from the database and performs bitwise operations to compare these codes with the matching string, calculating the Hamming distance. If the calculation result shows that the encoding string is highly similar to the biometric code of a family member, the system determines that the identity match is successful, returns the user's identity information, and selects the appropriate verification mode (such as fingerprint + facial dual verification) accordingly, ultimately completing the door lock unlocking operation.

[0025] In a specific scenario, the preliminary screening of the code string to be matched based on the feature index table to obtain candidate matching codes includes: The hash value of the encoding string to be matched is extracted to obtain the hash value to be matched, and the standard biometric codes in the feature index table are mapped by hash index block to obtain the hash index table; Based on the hash value to be matched, an approximate value lookup is performed on the hash index table to obtain a set of approximate hash values, and the approximate hash value set is encoded and traced to obtain a traceability code; The source code is partitioned for matching to obtain the partition matching result, and the source code is initially filtered based on the partition matching result to obtain the filtered code; The feature bit weights of the filtering codes are calculated to obtain the coding weight values, and the filtering codes are sorted and filtered based on the coding weight values ​​to obtain candidate matching codes.

[0026] Specifically, firstly, in the hash value extraction stage, the system performs hash calculations on the currently generated code string to be matched, extracting a hash value that represents the overall characteristics of the code string, i.e., the hash value to be matched. Simultaneously, the system performs a hash index block mapping operation on the standard biometric codes in the pre-stored user identity database, dividing all standard codes in the database into several hash index blocks according to their hash values, thereby constructing a structured hash index table. This hash index block mechanism not only improves database retrieval efficiency but also allows the system to maintain a high response speed when dealing with large-scale user data. Next, the system uses the hash value to be matched to perform an approximate value search in the hash index table to obtain a set of hash values ​​with similar characteristics to the current code string to be matched, i.e., the approximate hash value set. Since biometric data may be affected by environmental factors, posture, noise, etc., during the collection process, direct precise hash matching often fails to meet practical needs. Therefore, the system adopts an approximate matching strategy, allowing the search for similar hash values ​​within a certain error range. Subsequently, the system performs source tracing on these approximate hash values, that is, it traces back to the corresponding standard biometric codes according to the mapping relationship in the hash index table, thereby obtaining a set of preliminary matching candidate codes, i.e., source codes. After obtaining the source codes, the system further performs a partition matching operation, dividing the source codes into multiple feature regions according to their feature bit distribution, and performing local comparison with the code string to be matched in each region to generate partition matching results. This process aims to enhance the robustness of matching through local feature comparison and avoid overall misjudgment due to mismatches of certain feature bits. Based on the matching situation of each partition, the system performs preliminary filtering on the source codes, eliminating those codes that have significant differences in key feature regions, and retaining codes that are highly consistent with the code string to be matched in multiple feature regions, i.e., filtered codes. Subsequently, the system performs feature bit weight calculation on the filtered codes, that is, assigning different weight values ​​to each feature bit in each code to reflect its importance in identity recognition. For example, in fingerprint recognition, certain feature points (such as bifurcation points) may be more valuable for identification than others, thus their corresponding feature bits are assigned higher weights. Similarly, in facial recognition, features of key areas such as the corners of the eyes and the tip of the nose may have higher recognition weights. The system generates a weighted value for each filter code through weighted calculations, reflecting the credibility of the code's match with the current user's identity. Finally, the system sorts and filters the filter codes based on these weighted values, prioritizing codes with higher weights to generate the final candidate matching codes. These candidate codes will serve as inputs to the subsequent precise matching stage, participating in the final identity matching decision. Through this series of filtering mechanisms, the system can significantly reduce the number of codes participating in precise matching while ensuring recognition accuracy, thereby improving overall matching efficiency.For example, in a home application scenario, a user attempts to unlock the door using facial recognition. The system captures their facial image and generates a standardized biometric image. After feature extraction and encoding compression, a matching coded string is formed. The system extracts the hash value from this coded string and searches for a set of approximate hash values ​​in a hash index table, tracing back several source codes. Subsequently, the system performs partition matching and preliminary filtering on these codes, retaining filtered codes that match the current user's facial features well across multiple regions. Further feature bit weight calculations are performed on these codes, ultimately selecting candidate matching codes with higher weights. These candidate codes will then enter the subsequent precise matching stage, where bitwise operations are performed against the matching coded string to ultimately determine the user's identity.

[0027] In a specific scenario, the step of performing an approximate value lookup on the hash index table based on the hash value to be matched, to obtain a set of approximate hash values, includes: The hash value to be matched is divided by a sliding window to obtain several consecutive hash substrings to be matched, and each standard hash value in the hash index table is divided by the same window length to obtain an index hash substring group. The character overlap is calculated based on the hash substring to be matched and each substring in the index hash substring group to obtain the substring overlap value. The substring overlap value is then accumulated to obtain the total overlap between a single standard hash value and the hash value to be matched. The standard hash values ​​in the hash index table are initially screened by the total overlap to obtain candidate hash values ​​with a total overlap higher than a preset threshold. The candidate hash values ​​are then assigned feature bit weights to generate weighted candidate hash values. A comprehensive score is calculated based on the weighted candidate hash values ​​to obtain a hash value matching score. The weighted candidate hash values ​​are then sorted in descending order according to the hash value matching score to obtain an approximate hash value set.

[0028] Specifically, after completing hash value extraction and hash index table construction, the system enters the approximation search stage. At this point, the system first performs a sliding window segmentation on the hash value to be matched, dividing the hash value into several consecutive hash substrings to be matched according to a preset window length. For example, if the hash value to be matched is “11010110101011001101”, and the window length is set to 4, the system will sequentially extract hash substrings such as “1101”, “1010”, “0101”, “1010”, “0101”, “1011”, “0110”, “1100”, “1001”, “0011”, and “0110”. This sliding window segmentation method can effectively capture local features in the hash value, improving the robustness of subsequent matching. Simultaneously, the system also performs a segmentation operation of the same window length on each standard hash value stored in the hash index table, obtaining the corresponding index hash substring group. For example, a standard hash value of "11010110100011001100" will result in substrings containing information such as "1101", "1010", "0101", "1010", "0001", "0011", "0110", "1100", "1000", "0001", and "0011". This method ensures the consistency of feature segmentation between the hash value to be matched and the standard hash value, providing a structurally consistent data foundation for subsequent comparison and analysis. Next, the system calculates the character overlap between the hash substring to be matched and each substring in the index hash substring group to assess their similarity. Character overlap can be calculated by comparing each character individually and counting the number of identical characters, or by using more complex similarity measurement models such as Jaccard similarity or cosine similarity. For example, if a substring to be matched is "1101" and an index substring is "1111", their character overlap is 3 / 4, or 75%. After comparing each substring one by one, the system accumulates the overlap values ​​of each substring to obtain the total overlap between a single standard hash value and the hash value to be matched. This total overlap reflects the overall similarity between the standard hash value and the hash value to be matched. Subsequently, the system uses this total overlap to perform preliminary screening of all standard hash values ​​in the hash index table, retaining those candidate hash values ​​whose total overlap is higher than a preset threshold. For example, if the set overlap threshold is 80%, only standard hash values ​​whose total overlap reaches or exceeds this value will be retained as candidate matching objects. This preliminary screening mechanism can significantly narrow down the range of subsequent precise matching, thereby improving the overall recognition efficiency. To further improve the screening accuracy, the system assigns feature bit weights to these candidate hash values, generating weighted candidate hash values.The weighting of feature bits can be based on the importance of different bit segments in biometric coding. For example, in fingerprint recognition, certain feature points (such as bifurcation points) have higher discriminative power in identity recognition, so their corresponding hash bits can be assigned higher weights. The system uses a weighting mechanism to weight each bit segment of each candidate hash value, making it more representative in subsequent scoring. Finally, the system performs a comprehensive score calculation based on the weighted candidate hash values ​​to obtain a hash value matching score, and sorts all weighted candidate hash values ​​in descending order according to this score, ultimately generating an approximate hash value set. This score calculation can use weighted similarity, Euclidean distance, or other machine learning models for comprehensive evaluation to ensure the scientific validity and stability of the scoring results. For example, the system may score based on the number of bits differing between the weighted hash value and the hash value to be matched; a higher score indicates a higher degree of matching. Finally, the system sorts these hash values ​​from highest to lowest score to form an approximate hash value set for use in the subsequent precise matching stage. For example, in a typical home application scenario, a user attempts to unlock a door using facial recognition. The system captures their facial image and generates a standardized biometric image, then performs feature extraction and encoding compression to generate a hash value to be matched. The system performs a sliding window segmentation on this hash value and compares it with standard hash values ​​in a hash index table, calculating the total overlap. The system selects candidate hash values ​​with a total overlap higher than a threshold and performs feature bit weighting and comprehensive scoring calculations on them, ultimately generating a set of approximate hash values. The biometric codes corresponding to these hash values ​​will then enter the subsequent precise matching stage for final identity verification.

[0029] In a specific scenario, the step of performing bitwise operations to compare the candidate matching code with the code string to be matched, and generating a code difference value, includes: Each coding sequence in the candidate matching code is parsed by bit segment to obtain a hierarchical coding structure, and the hierarchical coding structure is position-calibrated to obtain a calibrated coding array; Based on the calibrated encoding array, a bit-by-bit XOR comparison is performed on the encoding string to be matched to obtain a bit difference comparison matrix, and the difference bit statistics are performed on the bit difference comparison matrix to obtain the difference bit distribution characteristics. Hamming distance quantization is performed on the differential bit distribution features to obtain distance metric parameters, and differential values ​​are calculated based on the distance metric parameters to obtain encoded differential values.

[0030] Specifically, candidate matching codes are several potentially matching biometric codes extracted from the user identity database during the initial screening stage. These codes share some structural similarity with the currently acquired matching code string, but have not yet undergone precise comparison and verification. To achieve more refined comparison, the system first performs bit-segment parsing on each coding sequence in the candidate matching codes, dividing it into multiple coding regions or levels with specific semantics. For example, fingerprint recognition may include hierarchical coding structures such as ridge distribution, feature point positions, and orientation information. This hierarchical structure helps the system differentiate between different feature dimensions during subsequent comparisons. After bit-segment parsing, the system further performs position calibration on these hierarchical coding structures to eliminate coding offsets caused by acquisition posture, angle, or device errors. For example, fingerprint images entered by a user at different times may have shifted feature point coding order due to different finger placement angles. The system aligns these codes through a position calibration mechanism to ensure they are logically consistent with the matching code string, thereby generating a calibrated coding array. This array serves as the basis for subsequent comparisons, effectively improving recognition accuracy. Next, the system performs a bit-by-bit XOR comparison of the calibrated encoding array with the encoding string to be matched, generating a bit difference comparison matrix. This process compares each bit of the two encoding strings, outputting 0 if they are the same and 1 if they are different, thus forming a binary matrix composed of 0s and 1s. This matrix intuitively reflects the distribution of differences between the candidate matching code and the encoding string to be matched, providing the raw data foundation for subsequent difference analysis. After obtaining the bit difference comparison matrix, the system performs difference bit statistics to extract the difference bit distribution characteristics. The difference bit distribution characteristics refer to the number of bits that differ during the comparison process and their distribution in the encoding string. For example, some regions have more feature bit differences, while other regions are highly consistent. Through statistical analysis of these distribution characteristics, the system can determine the degree of matching between the candidate matching code and the encoding string to be matched, and provide a basis for subsequent quantitative evaluation. Subsequently, the system performs Hamming distance quantization on the difference bit distribution characteristics to calculate the distance metric parameter. Hamming distance is a mathematical indicator that measures the degree of difference between two strings of equal length; its value is equal to the number of different characters at the same position in the two strings. In this scheme, the system calculates the Hamming distance between the candidate matching code and the code string to be matched, obtaining a quantitative parameter reflecting their similarity, namely the distance metric parameter. A higher parameter indicates a greater difference and a lower matching degree; conversely, a lower parameter indicates a higher matching degree. Finally, the system calculates the difference value based on this distance metric parameter, deriving the coding difference value. The coding difference value is a comprehensive evaluation index that considers not only the total number of differing bits but also factors such as feature bit weights, assigning different influence weights to differences in different feature regions.For example, in certain highly sensitive feature regions, even a small difference can significantly impact the identity recognition result. The system calculates the difference value to ultimately rank the candidate matching codes and selects the code closest to the target code string as the basis for identity matching. For instance, in a typical home scenario, a user attempts to unlock the door using facial recognition. The system captures their facial image and generates a standardized biometric image. After feature extraction and encoding compression, a target code string is formed. The system extracts candidate matching codes from the user identity database and performs bit segment parsing and position calibration to generate a calibrated code array. Subsequently, the system performs a bit-by-bit XOR comparison between the candidate codes and the target code string, generating a bit difference comparison matrix and statistically analyzing the distribution characteristics of the difference bits. The system further performs Hamming distance quantization on these features to obtain distance metric parameters and calculates the code difference value accordingly, ultimately determining whether the user is a family member and unlocking the door.

[0031] In a specific scenario, the step of performing a bit-by-bit XOR comparison of the encoded string to be matched based on the calibrated encoded array to obtain a bit difference comparison matrix includes: The calibrated encoded array is marked with feature bits to obtain a key location table, and the key location table is weighted to obtain a location weight sequence. Based on the position weight sequence, a region-specific XOR operation is performed on the encoding string to be matched to obtain a difference position map, and feature clustering is performed on the difference position map to obtain a difference clustering table. Spatial mapping is performed on the difference clustering table to obtain the difference localization matrix, and boundary correction is performed on the difference localization matrix to obtain the position difference comparison matrix.

[0032] Specifically, in this process, the bit-by-bit XOR comparison of the encoded string to be matched based on the calibrated encoding array to obtain the bit difference comparison matrix is ​​a key technical step in achieving high-precision identity recognition. This process first marks the calibrated encoding array with feature bits to obtain a key location table. This step means that the system, based on the biometric model established during pre-training or registration, identifies highly stable and discriminative bits in fingerprint ridge bifurcation points, facial key point contours, or iris textures, and records their location information to form a key location table. These locations typically correspond to the areas of biometrics least affected by changes in sampling posture, lighting, or pressure, such as the core triangle area in fingerprints or the symmetry axis area of ​​the bridge of the nose in the face, thus possessing stronger discriminative capabilities. Subsequently, the system assigns weights to the key location table to obtain a location weight sequence, that is, assigning corresponding weight values ​​according to the contribution of different feature bits to the overall recognition. For example, feature points located at the structural center with extremely low repetition rates are given higher weights, while edge positions or positions susceptible to noise interference have lower weights. This weighting mechanism makes subsequent comparisons more closely resemble the spatial distribution patterns of real biometric features. Next, based on the position weight sequence, a regional XOR operation is performed on the encoded string to be matched to obtain a difference position map. Specifically, the system divides the encoded string to be matched into several regions according to the same structure as the calibrated encoded array, and performs a bitwise XOR operation in each region. If two encoded values ​​are the same at a certain bit, the result is 0; otherwise, it is 1. This generates a binary map reflecting the distribution of bit differences—the difference position map. This map not only shows the locations where differences occur, but also prioritizes differences in key regions through the guidance of the position weight sequence. Following this, the system performs feature clustering on the difference position map to obtain a difference clustering table. This involves grouping adjacent or spatially continuous difference sites into one category, forming several difference clusters. For example, in face recognition, if multiple consecutive differences appear in the corner of the eye area, it may indicate that the user is wearing glasses or there is occlusion. The difference clustering table can effectively capture such local structural deviations rather than sporadic noise. Next, the system performs spatial mapping on the difference clustering table to obtain a difference localization matrix. This maps these difference clusters back to the coordinate space of the original biometric image, forming a difference heatmap matrix with spatial semantics, thus intuitively reflecting the specific distribution location of differences within the biometric structure. Finally, the system performs boundary correction on the difference localization matrix to obtain a bit difference comparison matrix. This step uses morphological processing or edge smoothing algorithms to eliminate irregular boundaries caused by acquisition errors or encoding distortion, ensuring more accurate geometric shapes of the difference regions. The final bit difference comparison matrix not only contains numerical information about the differences but also integrates spatial structure and weight distribution, providing a refined basis for subsequent identity matching decisions.For example, in a scenario where a family member comes home from get off work and uses facial recognition to unlock the door, the system captures their facial image and generates a matching encoding string. This string is then compared with the calibrated encoding array registered in the database. If the user was wearing a hat that day, resulting in missing information in the forehead area, the difference location map will show continuous differences in the corresponding area. After clustering and spatial mapping, the system identifies that the differences are concentrated in non-critical areas and do not affect high-weight locations such as the tip of the nose and pupils. Therefore, the system determines that the overall matching degree is still higher than the threshold, allowing the verification to be passed and the unlocking operation to be executed. This achieves intelligent recognition that is both accurate and fault-tolerant.

[0033] In a specific scenario, the step of selecting a verification mode for the door lock based on the identity matching result and generating the current verification mode includes: Based on the identity matching results, the user's security level is assessed to obtain the user security risk coefficient, and based on the user security risk coefficient, the permission level is calculated to obtain the user permission level parameter. Based on the user permission level parameters, a query is performed in the door lock's verification strategy library to obtain candidate verification modes, and the candidate verification modes are sorted by priority to obtain the current verification mode.

[0034] Specifically, the identity matching result is the user identity information obtained by the system after completing the biometric code comparison, including user ID, matching score, historical behavior records, and other data. Based on this information, the system first conducts a security level assessment of the user. The core of the assessment is to determine the credibility of the user's current identity verification and its potential security risks. For example, if a user's identity matching result shows that their biometric code is highly consistent with the standard code in the database, and the user is a long-term registered user, the system determines that their security risk coefficient is low; conversely, if there is a large discrepancy in the matching result, or if the user is a temporarily authorized user, the system will determine that they have a high security risk coefficient. This assessment process can be implemented by combining machine learning models, rule engines, or historical data analysis to ensure the scientific and real-time nature of the assessment results. After obtaining the user's security risk coefficient, the system further uses this coefficient to calculate permission levels and generate user permission level parameters. Permission level parameters are a quantitative representation of user access permissions, usually divided into multiple levels, such as high-level users (e.g., family members), mid-level users (e.g., long-term visitors), and low-level users (e.g., temporary visitors). Users with different permission levels will correspond to different verification strategies and access control rules. For example, advanced users may only need single-modal verification (such as fingerprint verification) to unlock the door, while less advanced users may need multiple verifications (such as fingerprint + facial recognition + dynamic password) to pass the verification. This hierarchical permission mechanism not only improves the system's security but also enhances the door lock's adaptability to different usage scenarios. Subsequently, based on the user's permission level parameter, the system queries the door lock's verification policy library to obtain candidate verification modes that match the permission level. The verification policy library is a set of preset verification rules, where each verification mode corresponds to a specific permission level and security scenario. For example, for advanced users, the system may configure fast verification modes such as "fingerprint verification" or "facial recognition + voice recognition"; while for less advanced users, it may configure high-security verification modes such as "fingerprint + facial recognition + SMS verification code". The system uses the permission level parameter as an index to retrieve a set of candidate verification modes that meet the current user's permissions and security requirements from the verification policy library. After obtaining the candidate verification modes, the system also needs to prioritize them to determine the final verification mode to be executed. Prioritization is typically based on multiple factors, including the current time, environmental security status, user behavior habits, and system resource load. For example, during the day or when family members frequently enter and exit, the system may prioritize modes that offer fast verification speeds and ease of use for the user; while at night or when the system detects abnormal behavior (such as multiple verification failures), it will increase the priority of verification modes and select more stringent verification methods to enhance security. Furthermore, the system can also consider users' historical preferences, prioritizing the verification methods most frequently used by the user to improve the user experience.For example, in a home application scenario, a family member returns home during the day and attempts to unlock the door using facial recognition. After identity verification, the system determines the user to be a high-level user with a low security risk factor and calculates the corresponding user permission level parameters. Based on these parameters, the system queries the verification policy library for two candidate verification modes: "facial recognition" and "fingerprint + facial recognition." Combining the current time (daytime) and environmental security status (no abnormal records), the system prioritizes these two modes and ultimately selects "facial recognition" as the current verification mode. After the user completes facial recognition, the system unlocks the door. The entire process is efficient, convenient, and secure. In summary, this solution constructs a dynamic, intelligent, and secure identity verification mechanism through steps such as security level assessment based on identity matching results, permission-level calculation, verification policy library query, and priority ranking of candidate verification modes. The entire process closely integrates user identity information and system policies, achieving precise control over the door lock verification mode. This not only improves the system's security and flexibility but also provides reliable verification support and user experience assurance for various smart door lock application scenarios such as homes, apartments, and offices.

[0035] In a specific scenario, the step of controlling the door lock state based on the current verification mode to complete the door lock state control operation includes: The current verification mode is mapped to control permissions to obtain a door lock operation permission code, and the door lock operation permission code is verified for security level to obtain an authorized execution signal; Based on the authorized execution signal, the lock cylinder drive mechanism of the door lock is controlled to obtain the lock cylinder rotation control sequence; Based on the lock cylinder rotation control sequence, the torque of the lock cylinder drive mechanism is adjusted to obtain the lock cylinder action execution parameters; The mechanical state data of the door lock is analyzed by the lock cylinder action execution parameters, and the opening and closing status is identified based on the mechanical state data to obtain the current status identifier of the door lock; The control result is confirmed based on the current status identifier of the door lock, a status control signal is obtained, and the door lock status control operation is completed based on the status control signal.

[0036] Specifically, the current verification mode is the user authentication method determined by the system after prior identity matching and permission assessment, such as fingerprint verification or dual verification of face and fingerprint. Based on this verification mode and pre-defined permission mapping rules, the system maps the verification mode to the door lock's control permissions, generating a door lock operation permission code. This permission code represents the user's door lock operation permissions under the current verification mode, such as only allowing unlocking, allowing temporary authorization, or prohibiting operation. Subsequently, the system performs a security level verification on the door lock operation permission code, checking its legality, validity, and whether it matches the current system state. For example, if the permission code is a temporary authorization code and has expired, the system will refuse to execute subsequent operations. Only if the verification passes will the system generate an authorization execution signal as the trigger condition for subsequent door lock operations. After obtaining the authorization execution signal, the system controls the door lock's lock cylinder drive mechanism based on this signal, generating a lock cylinder rotation control sequence. The lock cylinder drive mechanism typically consists of a motor, electromagnetic lock, or stepper motor. Its core function is to drive the lock cylinder to rotate or displace according to control signals, thereby opening or closing the door lock. Based on the authorized execution signal, the system generates a corresponding lock cylinder rotation control sequence, such as "rotate 90 degrees clockwise" or "rotate 45 degrees counterclockwise," and sends it to the drive module for execution. During the drive process, the system also needs to adjust the torque of the lock cylinder drive mechanism based on the lock cylinder rotation control sequence to adapt to mechanical resistance and user operating habits in different environments. For example, in winter, the door lock may experience increased friction due to the contraction of mechanical parts caused by low temperatures; in this case, the system needs to increase the drive torque to ensure the lock cylinder can rotate smoothly. The system obtains the real-time operating status of the lock cylinder drive mechanism through a feedback mechanism and, combined with a preset adjustment algorithm, generates lock cylinder action execution parameters. These parameters include drive voltage, current, rotation angle, and speed, ensuring the smoothness and reliability of the lock cylinder's operation. Simultaneously, the system analyzes the door lock's mechanical state data through the lock cylinder action execution parameters, such as whether the lock cylinder is in position, whether there is jamming, or whether there is abnormal vibration. These data are collected in real time by built-in sensors (such as angle sensors, pressure sensors, and vibration sensors) and analyzed by the system. Based on these analysis results, the system can identify the current open / closed state of the door lock, such as "fully open," "not fully open," or "closed," and generate a current door lock status identifier. This identifier is used not only to confirm the result of the current control operation but also for system log recording and anomaly detection. After obtaining the current door lock status identifier, the system confirms the control result and generates a status control signal. For example, if the system identifies that the door lock has been successfully opened, it generates a "successful unlock" status control signal and updates the system status; if it identifies that the door lock is not fully open or there is an anomaly, it generates a "failed unlock" status control signal and may trigger an alarm mechanism or prompt the user to perform the operation again.Ultimately, the system completes door lock status control operations based on status control signals, such as performing unlocking, locking, or entering a secure locked state. For example, in a home scenario, a user completes identity verification via facial recognition. The system generates a "facial recognition" verification mode based on the user's permission level and maps it to the corresponding door lock operation permission code. After performing a security check, the system generates an authorization execution signal and controls the lock cylinder drive mechanism to generate a lock cylinder rotation control sequence. The system dynamically adjusts the drive torque based on ambient temperature and lock cylinder feedback data to ensure smooth lock cylinder rotation and collects mechanical status data through sensors to determine whether the door lock is fully open. Finally, the system generates a status control signal based on the recognition result, completes the unlocking operation, and sends a "successfully unlocked" status feedback to the user via the app.

[0037] The above describes the smart lock security control method with multi-mode switching in the embodiments of the present invention. The following describes the smart lock security control device with multi-mode switching in the embodiments of the present invention. Please refer to [link / reference]. Figure 2 One embodiment of the multi-mode switching smart door lock security control device of the present invention includes: The acquisition module 21 is used to acquire the user's biometric data and perform feature encoding on the user's biometric data to obtain a biometric encoding sequence; The calculation module 22 is used to perform matching calculations between the biometric encoding sequence and the pre-stored user identity database to obtain the identity matching result; Selection module 23 is used to select the verification mode of the door lock based on the identity matching result and generate the current verification mode; Control module 24 is used to control the state of the door lock based on the current verification mode and complete the door lock state control operation.

[0038] In this embodiment, the specific implementation of each unit in the above device embodiment is described in the above method embodiment, and will not be repeated here.

[0039] Reference Figure 3 This invention provides a virtual human lecturing video generation device, comprising: At least one processor 301; At least one memory 302 is used to store at least one program; When at least one program is executed by at least one processor 301, the at least one processor 301 implements a method for generating virtual human lecture videos.

[0040] Similarly, the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0041] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this invention are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is altered and sub-operations described as part of a larger operation are executed independently.

[0042] Furthermore, although the invention has been described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the described functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the invention. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of conventional skill of an engineer. Therefore, those skilled in the art can implement the invention as set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the specific concepts disclosed are merely illustrative and not intended to limit the scope of the invention, which is determined by the full scope of the appended claims and their equivalents.

[0043] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several programs to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0044] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequential list of executable programs for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, a program execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can retrieve and execute a program from or in conjunction with such a program execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can mean any means that can contain, store, communicate, propagate, or transmit a program for use by or in conjunction with a program execution system, apparatus, or device.

[0045] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0046] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable program execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0047] In the foregoing description of this specification, references to terms such as "one embodiment," "another embodiment," or "some embodiments" indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of the present invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0048] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

[0049] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.

Claims

1. A smart door lock security control method with multi-mode switching, characterized in that, Includes the following steps: Collect user biometric data and perform feature encoding on the user biometric data to obtain a biometric encoding sequence; The biometric encoding sequence is matched with a pre-stored user identity database to obtain the identity matching result; Based on the identity matching result, the door lock is selected for verification mode, and the current verification mode is generated. Based on the current verification mode, the door lock is state controlled to complete the door lock state control operation.

2. The multi-mode switching smart door lock security control method according to claim 1, characterized in that, The process of collecting user biometric data and encoding the user biometric data to obtain a biometric encoding sequence includes: The user's original biometric data is collected by an identity recognition module installed on the door lock, and noise suppression processing is performed on the original biometric data to obtain a standardized biometric image, which is then used as the user's biometric data. The distribution of local feature points is extracted from the user's biometric data, feature point coordinates are generated, and geometric invariant transformation and normalization are performed on the feature point coordinates to generate a feature description vector at a uniform scale. Binarization encoding and bit string compression operations are performed on each vector in the feature description vector to obtain the biometric feature encoding sequence.

3. The multi-mode switching smart door lock security control method according to claim 1, characterized in that, The step of matching the biometric encoded sequence with a pre-stored user identity database to obtain an identity matching result includes: Obtain the feature index table corresponding to the standard biometric code that is pre-stored in the pre-stored user identity database, and convert the format of the biometric code sequence to obtain the code string to be matched; Based on the feature index table, the string to be matched is initially screened to obtain candidate matching codes, and bitwise operations are performed on the candidate matching codes and the string to be matched to generate a bitwise comparison result. The user's identity matching result is determined by comparing the encoded bits.

4. The multi-mode switching smart door lock security control method according to claim 3, characterized in that, The preliminary screening of the code string to be matched based on the feature index table to obtain candidate matching codes includes: The hash value of the encoding string to be matched is extracted to obtain the hash value to be matched, and the standard biometric codes in the feature index table are mapped by hash index block to obtain the hash index table; Based on the hash value to be matched, an approximate value lookup is performed on the hash index table to obtain a set of approximate hash values, and the approximate hash value set is encoded and traced to obtain a traceability code; The source code is partitioned for matching to obtain the partition matching result, and the source code is initially filtered based on the partition matching result to obtain the filtered code; The feature bit weights of the filtering codes are calculated to obtain the coding weight values, and the filtering codes are sorted and filtered based on the coding weight values ​​to obtain candidate matching codes.

5. The multi-mode switching smart door lock security control method according to claim 3, characterized in that, The step of comparing the candidate matching code with the code string to be matched by bitwise operations to generate a code difference value includes: Each coding sequence in the candidate matching code is parsed by bit segment to obtain a hierarchical coding structure, and the hierarchical coding structure is position-calibrated to obtain a calibrated coding array; Based on the calibrated encoding array, a bit-by-bit XOR comparison is performed on the encoding string to be matched to obtain a bit difference comparison matrix, and the difference bit statistics are performed on the bit difference comparison matrix to obtain the difference bit distribution characteristics. Hamming distance quantization is performed on the differential bit distribution features to obtain distance metric parameters, and differential values ​​are calculated based on the distance metric parameters to obtain encoded differential values.

6. The multi-mode switching smart door lock security control method according to claim 1, characterized in that, The step of selecting a verification mode for the door lock based on the identity matching result and generating the current verification mode includes: Based on the identity matching results, the user's security level is assessed to obtain the user security risk coefficient, and based on the user security risk coefficient, the permission level is calculated to obtain the user permission level parameter. Based on the user permission level parameters, a query is performed in the door lock's verification strategy library to obtain candidate verification modes, and the candidate verification modes are sorted by priority to obtain the current verification mode.

7. The multi-mode switching smart door lock security control method according to claim 1, characterized in that, The step of controlling the door lock state based on the current verification mode to complete the door lock state control operation includes: The current verification mode is mapped to control permissions to obtain a door lock operation permission code, and the door lock operation permission code is verified for security level to obtain an authorized execution signal; Based on the authorized execution signal, the lock cylinder drive mechanism of the door lock is controlled to obtain the lock cylinder rotation control sequence; Based on the lock cylinder rotation control sequence, the torque of the lock cylinder drive mechanism is adjusted to obtain the lock cylinder action execution parameters; The mechanical state data of the door lock is analyzed by the lock cylinder action execution parameters, and the opening and closing status is identified based on the mechanical state data to obtain the current status identifier of the door lock; The control result is confirmed based on the current status identifier of the door lock, a status control signal is obtained, and the door lock status control operation is completed based on the status control signal.

8. A multi-mode switching intelligent door lock security control device, characterized in that, include: The acquisition module is used to collect the user's biometric data and perform feature encoding on the user's biometric data to obtain a biometric encoding sequence; The calculation module is used to perform matching calculations between the biometric encoding sequence and the pre-stored user identity database to obtain the identity matching result; The selection module is used to select the verification mode of the door lock based on the identity matching result and generate the current verification mode. The control module is used to control the state of the door lock based on the current verification mode and complete the door lock state control operation.

9. A multi-mode switching intelligent door lock security control device, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements a multi-mode switching smart door lock security control method as described in any one of claims 1-7.

10. A storage medium storing a processor-executable program, characterized in that, The processor-executable program, when executed by the processor, is used to implement a multi-mode switching smart door lock security control method as described in any one of claims 1-7.

Images