Face recognition method and system for intelligent door lock based on posture angle and intelligent door lock

Abstract

The application discloses a face recognition method and system of an intelligent door lock based on a posture angle and an intelligent door lock, and the method comprises the following steps: acquiring a face image of a current user and a current time when a face is detected, acquiring a historical time when a user last closed a door lock according to historical door opening and closing record data, calculating a time interval, judging whether the time interval is less than a preset time threshold, if the time interval is less than the preset time threshold, confirming a face posture according to the face image, obtaining a roll angle, a pitch angle and a yaw angle, judging whether the roll angle, the pitch angle and the yaw angle simultaneously satisfy a preset angle range condition, if yes, confirming whether the face image of the current user is a face of a registered user, if yes, opening the door lock, and voice prompting that the door lock has been opened. The application improves the accuracy of the intelligent door lock recognition without affecting the normal use of the user, reduces the mislocking condition, and ensures that the door lock is only opened when the user has an intention and meets specific safety conditions.

Description

Technical Field

[0001] This invention relates to the field of smart lock technology, and in particular to a smart lock face recognition method, system, smart lock, and computer-readable storage medium based on attitude angle. Background Technology

[0002] With the development of smart home technology, smart door locks have become an important part of modern homes. Existing smart door locks generally use facial recognition technology for convenient unlocking. However, this technology has a security issue: when a user looks back or adjusts their posture after opening the door, they may accidentally trigger the facial unlocking module, causing the lock to be opened incorrectly without the user's knowledge, thus leading to a security problem.

[0003] Existing facial recognition technology may not be able to accurately distinguish whether a user intentionally wants to unlock the door or accidentally triggers the unlocking action, especially in the short period of time after the user opens the door.

[0004] Due to the possibility of misunderstandings, users may have doubts about the reliability of smart locks, which could affect their trust in smart locks and their overall user experience.

[0005] Therefore, existing technologies still need to be improved and developed. Summary of the Invention

[0006] The main objective of this invention is to provide a face recognition method, system, smart lock, and computer-readable storage medium for smart locks based on attitude angles. This aims to solve the problem that existing smart locks cannot accurately distinguish whether a user intends to unlock the door or accidentally triggers the unlocking action. If the unlocking is accidentally triggered, the door lock may be opened incorrectly, causing security risks.

[0007] To achieve the above objectives, the present invention provides a face recognition method for smart locks based on posture angles, the face recognition method for smart locks based on posture angles comprising the following steps:

[0008] When a face is detected, the current user's face image and the current time are obtained. Based on the historical door opening and closing record data, the historical time of the most recent time when the user closed the door lock is obtained. The time interval is calculated based on the current time and the historical time.

[0009] Determine whether the time interval is less than a preset time threshold. If the time interval is less than the preset time threshold, confirm the face posture based on the current user's face image to obtain the flip angle, pitch angle, and yaw angle.

[0010] Determine whether the roll angle, pitch angle, and yaw angle simultaneously meet the preset angle range conditions. If they do, confirm whether the current user's face image is a registered user's face. If it is a registered user's face, unlock the door and provide a voice prompt that the door is unlocked.

[0011] Optionally, the smart door lock face recognition method based on attitude angle, wherein when a face is detected, the current user's face image and the current time are acquired, the historical time of the most recent time the user closed the door is obtained based on historical door opening and closing records, and the time interval is calculated based on the current time and the historical time, specifically includes:

[0012] Acquire the current user's facial image captured by the depth camera, and obtain the current time;

[0013] The face image is preprocessed, including noise reduction, contrast enhancement, color correction, and image normalization.

[0014] Obtain historical door opening and closing records, and based on the historical door opening and closing records, obtain the historical time of the most recent time when the user closed the door lock. Calculate the time interval based on the current time and the historical time.

[0015] Optionally, the smart door lock face recognition method based on attitude angle, wherein determining whether the time interval is less than a preset time threshold, and if the time interval is less than the preset time threshold, confirming the face attitude based on the current user's face image to obtain the flip angle, pitch angle, and yaw angle, specifically includes:

[0016] A preset time threshold is set in advance, and it is determined whether the time interval is less than the preset time threshold.

[0017] If the time interval is less than the preset time threshold, the face region in the face image is determined using a face detection algorithm;

[0018] The key feature points of a face are located using a feature point detection algorithm, and the roll angle, pitch angle and yaw angle in the attitude parameters are determined based on the key feature points.

[0019] Optionally, the smart door lock face recognition method based on attitude angle further includes, after determining whether the time interval is less than a preset time threshold:

[0020] If the time interval is greater than or equal to the preset time threshold, then it is determined whether the current user's face image is a registered user's face;

[0021] If the current user's facial image is that of a registered user, then unlock the door.

[0022] If the current user's face image is not a registered user's face, the door lock will not be opened.

[0023] Optionally, the smart door lock face recognition method based on attitude angle, wherein determining whether the flip angle, the pitch angle, and the yaw angle simultaneously satisfy a preset angle range condition, and if they simultaneously satisfy the preset angle range condition, confirming whether the current user's face image is a registered user's face, and if it is a registered user's face, unlocking the door and providing a voice prompt that the door lock is unlocked, specifically includes:

[0024] A preset angle range condition is set in advance, wherein the angle change range is within a preset angle, and it is determined whether the change ranges of the roll angle, the pitch angle and the yaw angle are all within the preset angle;

[0025] If the range of change of the flip angle, the pitch angle and the yaw angle are all within the preset angle, it indicates that the current user is facing the door lock. Then it is determined whether the face image of the current user is a registered user's face.

[0026] If the current user's face image is identified as that of a registered user, the door lock will be opened, and a voice prompt will be given to the current user indicating that the door lock is open.

[0027] Optionally, the smart door lock face recognition method based on attitude angle further includes, after determining whether the range of change of the flip angle, the pitch angle, and the yaw angle are all within a preset angle:

[0028] If the range of change of any one of the roll angle, pitch angle and yaw angle exceeds the preset angle, it indicates that the current user is not facing the door lock, and the door lock will not be opened.

[0029] Optionally, the smart door lock face recognition method based on pose angle further includes, after determining whether the current user's face image is a registered user's face:

[0030] If the current user's face image is not identified as belonging to a registered user, the door lock will not be opened.

[0031] Furthermore, to achieve the above objectives, the present invention also provides a smart door lock face recognition system based on attitude angle, wherein the smart door lock face recognition system based on attitude angle:

[0032] The face recognition module is used to acquire the current user's face image and the current time when a face is detected, obtain the historical time of the most recent time the user closed the door lock based on historical door opening and closing records, and calculate the time interval based on the current time and the historical time.

[0033] The attitude detection module is used to determine whether the time interval is less than a preset time threshold. If the time interval is less than the preset time threshold, the facial attitude is confirmed based on the current user's facial image, and the flip angle, pitch angle and yaw angle are obtained.

[0034] The door lock control module is used to determine whether the roll angle, pitch angle and yaw angle simultaneously meet the preset angle range conditions. If they simultaneously meet the preset angle range conditions, it confirms whether the current user's face image is a registered user's face. If it is a registered user's face, the door lock is opened and a voice prompt is given that the door lock is open.

[0035] Furthermore, to achieve the above objectives, the present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a smart door lock face recognition program based on attitude angle, and the smart door lock face recognition program based on attitude angle, when executed by a processor, implements the steps of the smart door lock face recognition method based on attitude angle as described above.

[0036] In this invention, when a face is detected, the current user's facial image and the current time are acquired. Based on historical door opening and closing records, the most recent time the user closed the door is obtained. A time interval is calculated based on the current time and the historical time. It is then determined whether the time interval is less than a preset time threshold. If the time interval is less than the preset time threshold, the facial posture is confirmed based on the current user's facial image, obtaining the flip angle, pitch angle, and yaw angle. It is then determined whether the flip angle, pitch angle, and yaw angle simultaneously meet preset angle range conditions. If they do, it is confirmed whether the current user's facial image is that of a registered user. If it is, the door is unlocked, and a voice prompt indicates that the door is unlocked. This invention improves the accuracy of smart door lock recognition without affecting normal user operation, reduces false locking, and ensures that the door is only unlocked when the user intends to and meets specific security conditions, thus improving security. Attached Figure Description

[0037] Figure 1 This is a flowchart of a preferred embodiment of the smart door lock face recognition method based on attitude angle of the present invention;

[0038] Figure 2 This is a schematic diagram of the flip angle, pitch angle and yaw angle in a preferred embodiment of the smart door lock face recognition method based on attitude angle of the present invention;

[0039] Figure 3 This is a schematic diagram illustrating the entire face recognition and door lock control process in a preferred embodiment of the intelligent door lock face recognition method based on attitude angle of the present invention.

[0040] Figure 4 This is a schematic diagram illustrating the principle of a preferred embodiment of the intelligent door lock face recognition system based on attitude angle of the present invention;

[0041] Figure 5 This is a schematic diagram of the composition of a health data monitoring device in a preferred embodiment of the intelligent door lock face recognition system based on attitude angle of the present invention. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0043] The preferred embodiment of the present invention describes a smart door lock face recognition method based on attitude angle, such as... Figure 1 As shown, the smart door lock face recognition method based on attitude angle includes the following steps:

[0044] Step S10: When a face is detected, obtain the current user's face image and the current time, obtain the historical time of the most recent time the user closed the door lock based on the historical door opening and closing record data, and calculate the time interval based on the current time and the historical time.

[0045] Specifically, the smart door lock of the present invention can use a high-definition camera to capture the user's facial image. In order to obtain more accurate posture information, it may be necessary to use multiple cameras to acquire images from different angles, or to use a depth camera to acquire three-dimensional information. For example, the depth camera can capture the current user's facial image and obtain the current time. The facial image is preprocessed, including noise reduction, contrast enhancement, color correction, and image normalization, which can improve the accuracy of subsequent image processing. Then, historical door opening and closing record data is obtained. Based on the historical door opening and closing record data, the historical time of the most recent time when the user closed the door lock (indicating that the user closed the door lock when leaving the house) is obtained. The time interval is calculated based on the current time and the historical time.

[0046] Step S20: Determine whether the time interval is less than a preset time threshold. If the time interval is less than the preset time threshold, confirm the face posture based on the current user's face image and obtain the flip angle, pitch angle, and yaw angle.

[0047] Specifically, a preset time threshold is set (e.g., the preset time threshold can be set to 5 seconds, which can be adjusted according to the user's actual needs). The system determines whether the time interval is less than the preset time threshold. If the time interval is greater than or equal to the preset time threshold, it identifies whether the current user's face image is that of a registered user. If the current user's face image is that of a registered user, the door lock is opened; if the current user's face image is not that of a registered user, the door lock is not opened. If the time interval is less than the preset time threshold, a face detection algorithm (e.g., Haar cascade, deep learning model, etc.) is used to determine the face region in the face image (the region that facilitates subsequent pose analysis). A feature point detection algorithm (e.g., AdaBoost, LBP, or deep learning model) is used to locate key facial feature points (e.g., eyes, nose, mouth, and cheeks). Based on the key feature points, the roll angle, pitch angle, and yaw angle in the pose parameters are determined. Figure 2 As shown, this can be achieved through geometric modeling, machine learning algorithms, or deep learning methods. For example, a 3D face model can be used to estimate the pose, or a pose estimation network (such as PRNet, FaceNet, etc.) can be used to estimate the pose directly from a 2D image.

[0048] Furthermore, the roll angle, pitch angle, and yaw angle in the attitude parameters can be determined in various ways, for example:

[0049] (1) Facial landmark detection: Facial landmark detection algorithm is a commonly used face detection algorithm that can detect the location of key points in a face image, such as the positions of the eyes, mouth, and nose. By detecting the position of these key points, the pose angle of the face can be determined. For example, by detecting the position and distance of the eyes, the pitch and yaw angles of the face can be estimated; by detecting the position of the mouth and nose, the flip angle of the face can be estimated.

[0050] (2) Face Pose Estimation: Face pose estimation algorithms are based on facial landmark detection. They can estimate the pose angle of a face based on the position of facial landmarks and the geometry of the face. For example, by calculating the position and distance of the eyes, the pitch and yaw angles of the face can be estimated; by calculating the position of the mouth and nose, the flip angle of the face can be estimated. In addition, face pose estimation algorithms can also use other features and deep learning models to improve accuracy.

[0051] (3) 3D Face Reconstruction: A 3D face reconstruction algorithm is an algorithm that converts a face image into a 3D model, which can be used to determine the pose angle of the face. For example, a 3D face model can be built, the face image can be mapped onto the 3D model, and then the 3D model can be used to calculate the pose angle of the face.

[0052] (4) Deep learning models: Deep learning models are neural network-based models that can classify and regress face images to obtain the pose angle of the face. For example, convolutional neural networks can be used to classify face images to determine the pose angle of the face; regression models can be used to regress face images to obtain the specific pose angle of the face.

[0053] In summary, all of the above methods can be used to determine the pose angle of a face image, but each method has its own advantages, disadvantages, and applicable scope. The appropriate method should be selected based on the specific application scenario and requirements.

[0054] Step S30: Determine whether the roll angle, pitch angle and yaw angle simultaneously meet the preset angle range conditions. If they simultaneously meet the preset angle range conditions, confirm whether the current user's face image is a registered user's face. If it is a registered user's face, unlock the door and give a voice prompt that the door lock is unlocked.

[0055] Specifically, a preset angle range condition is set in advance. This preset angle range condition means that the angle change range is within a preset angle (e.g., the change range is limited to ±15 degrees, which can be set according to the user's actual needs). It is determined whether the change ranges of the roll angle, pitch angle, and yaw angle are all within the preset angle. If the change range of any one of the roll angle, pitch angle, and yaw angle exceeds the preset angle, it indicates that the current user is not directly facing the door lock (e.g., the user only turns around to check if the door is closed properly or to adjust their posture after closing it), and the door lock is not opened. If the roll angle... If the pitch angle and yaw angle are both within a preset range, it indicates that the current user is facing the door lock (meaning the current user may need to unlock the door). Then, it is determined whether the current user's face image is a registered user's face. If the current user's face image is not a registered user's face, the door lock is not unlocked. If the current user's face image is a registered user's face, the door lock is unlocked, and a voice prompt is given to the current user that the door lock is unlocked (enhancing user experience). This effectively solves the problem of users accidentally unlocking the door without realizing it when looking back or adjusting their posture after closing it.

[0056] Furthermore, such as Figure 3 As shown, the process of the smart door lock face recognition method based on posture angle is as follows:

[0057] Step S101: The face unlock module in the smart door lock is triggered;

[0058] Step S102: The smart door lock captures the current user's facial image through a depth camera and obtains the current time;

[0059] Step S103: Obtain the historical time of the most recent time the user closed the door lock based on the historical door opening and closing record data, and calculate the time interval based on the current time and the historical time;

[0060] Step S104: Determine whether the time interval is less than a preset time threshold;

[0061] Step S105: If the time interval is greater than or equal to the preset time threshold, then identify whether the current user's face image is a registered user's face. If so, then open the door lock; otherwise, do not open the door lock.

[0062] Step S106: If the time interval is less than the preset time threshold, the face pose in the face image is confirmed based on the current user face image, and the flip angle, pitch angle and yaw angle are obtained.

[0063] Step S107: Determine whether the roll angle, pitch angle and yaw angle meet the preset angle range conditions; if they do not meet the preset angle range conditions, do not unlock the door and return to step S102.

[0064] Step S108: If the preset angle range conditions are met, further confirm whether the current user's face image is the face of a registered user. If so, unlock the door and give the user a voice prompt that the door is unlocked (the reason for adding the voice prompt is to remind the user that if the user turns around and the position and angle unexpectedly meet the preset conditions, the door will be opened without the user's knowledge. Therefore, the voice prompt is added to improve the user experience). Otherwise, do not unlock the door.

[0065] The smart door lock face recognition method based on posture angle provided by this invention effectively prevents users from accidentally triggering the face unlock function due to turning around or adjusting their posture after opening the door by limiting the recognition angle of the face recognition module and introducing a time threshold, thereby improving the security and reliability of the smart door lock.

[0066] Furthermore, such as Figure 4 As shown, based on the above-mentioned smart door lock face recognition method based on posture angle, the present invention also provides a smart door lock face recognition system based on posture angle, wherein the smart door lock face recognition system based on posture angle includes:

[0067] The face recognition module 51 is used to acquire the current user's face image and the current time when a face is detected, acquire the historical time of the most recent time the user closed the door lock based on historical door opening and closing records, and calculate the time interval based on the current time and the historical time.

[0068] The attitude detection module 52 is used to determine whether the time interval is less than a preset time threshold. If the time interval is less than the preset time threshold, the face attitude is confirmed based on the current user's face image, and the flip angle, pitch angle and yaw angle are obtained.

[0069] The door lock control module 53 is used to determine whether the roll angle, the pitch angle and the yaw angle simultaneously meet the preset angle range conditions. If they simultaneously meet the preset angle range conditions, it confirms whether the current user's face image is a registered user's face. If it is a registered user's face, it unlocks the door and provides a voice prompt that the door lock is unlocked.

[0070] Furthermore, such as Figure 5 As shown, based on the above-mentioned smart door lock face recognition method and system based on attitude angle, the present invention also provides a smart door lock, which includes a processor 10, a memory 20 and a display 30. Figure 5 Only some components of the smart lock are shown; however, it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0071] In some embodiments, the memory 20 may be an internal storage unit of the smart lock, such as a hard drive or memory. In other embodiments, the memory 20 may be an external storage device of the smart lock, such as a plug-in hard drive, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the smart lock. Further, the memory 20 may include both internal and external storage units of the smart lock. The memory 20 is used to store application software and various types of data installed on the smart lock, such as the program code for installing the smart lock. The memory 20 can also be used to temporarily store data that has been output or will be output. In one embodiment, the memory 20 stores a smart lock face recognition program 40 based on attitude angles, which can be executed by the processor 10 to implement the smart lock face recognition method based on attitude angles in this application.

[0072] In some embodiments, the processor 10 may be a central processing unit (CPU), a microprocessor, or other data processing chip, used to run program code stored in the memory 20 or process data, such as executing the smart door lock face recognition method based on attitude angle.

[0073] In some embodiments, the display 30 may be an LED display, a liquid crystal display, a touch-screen liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. The display 30 is used to display information about the smart lock and to display a visual user interface. The components 10-30 of the smart lock communicate with each other via a system bus.

[0074] In one embodiment, when the processor 10 executes the smart door lock face recognition program 40 based on attitude angle in the memory 20, the following steps are performed:

[0075] When a face is detected, the current user's face image and the current time are obtained. Based on the historical door opening and closing record data, the historical time of the most recent time when the user closed the door lock is obtained. The time interval is calculated based on the current time and the historical time.

[0076] Determine whether the time interval is less than a preset time threshold. If the time interval is less than the preset time threshold, confirm the face posture based on the current user's face image to obtain the flip angle, pitch angle, and yaw angle.

[0077] Determine whether the roll angle, pitch angle, and yaw angle simultaneously meet the preset angle range conditions. If they do, confirm whether the current user's face image is a registered user's face. If it is a registered user's face, unlock the door and provide a voice prompt that the door is unlocked.

[0078] Specifically, when a face is detected, the system acquires the current user's facial image and the current time, obtains the most recent time the user closed the door lock based on historical door opening and closing records, and calculates the time interval based on the current time and the historical time.

[0079] Acquire the current user's facial image captured by the depth camera, and obtain the current time;

[0080] The face image is preprocessed, including noise reduction, contrast enhancement, color correction, and image normalization.

[0081] Obtain historical door opening and closing records, and based on the historical door opening and closing records, obtain the historical time of the most recent time when the user closed the door lock. Calculate the time interval based on the current time and the historical time.

[0082] Specifically, determining whether the time interval is less than a preset time threshold, and if the time interval is less than the preset time threshold, confirming the facial pose based on the current user's facial image to obtain the flip angle, pitch angle, and yaw angle, includes:

[0083] A preset time threshold is set in advance, and it is determined whether the time interval is less than the preset time threshold.

[0084] If the time interval is less than the preset time threshold, the face region in the face image is determined using a face detection algorithm;

[0085] The key feature points of a face are located using a feature point detection algorithm, and the roll angle, pitch angle and yaw angle in the attitude parameters are determined based on the key feature points.

[0086] The step of determining whether the time interval is less than a preset time threshold further includes:

[0087] If the time interval is greater than or equal to the preset time threshold, then it is determined whether the current user's face image is a registered user's face;

[0088] If the current user's facial image is that of a registered user, then unlock the door.

[0089] If the current user's face image is not a registered user's face, the door lock will not be opened.

[0090] Specifically, determining whether the roll angle, pitch angle, and yaw angle simultaneously meet a preset angle range condition; if they do, confirming whether the current user's face image is a registered user's face; if it is a registered user's face, unlocking the door and providing a voice prompt that the door is unlocked; this includes:

[0091] A preset angle range condition is set in advance, wherein the angle change range is within a preset angle, and it is determined whether the change ranges of the roll angle, the pitch angle and the yaw angle are all within the preset angle;

[0092] If the range of change of the flip angle, the pitch angle and the yaw angle are all within the preset angle, it indicates that the current user is facing the door lock. Then it is determined whether the face image of the current user is a registered user's face.

[0093] If the current user's face image is identified as that of a registered user, the door lock will be opened, and a voice prompt will be given to the current user indicating that the door lock is open.

[0094] The step of determining whether the ranges of change of the roll angle, the pitch angle, and the yaw angle are all within a preset angle includes:

[0095] If the range of change of any one of the roll angle, pitch angle and yaw angle exceeds the preset angle, it indicates that the current user is not facing the door lock, and the door lock will not be opened.

[0096] The step of determining whether the current user's facial image is that of a registered user further includes:

[0097] If the current user's face image is not identified as belonging to a registered user, the door lock will not be opened.

[0098] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a smart door lock face recognition program based on attitude angle, and the smart door lock face recognition program based on attitude angle, when executed by a processor, implements the steps of the smart door lock face recognition method based on attitude angle as described above.

[0099] In summary, this invention provides a smart door lock face recognition method and system based on attitude angles. The method includes: acquiring the current user's face image and current time when a face is detected; acquiring the most recent time the user closed the door based on historical door opening and closing records; calculating a time interval based on the current time and the historical time; determining whether the time interval is less than a preset time threshold; if the time interval is less than the preset time threshold, confirming the face attitude based on the current user's face image to obtain the flip angle, pitch angle, and yaw angle; determining whether the flip angle, pitch angle, and yaw angle simultaneously satisfy a preset angle range condition; if they simultaneously satisfy the preset angle range condition, confirming whether the current user's face image is a registered user's face; if it is a registered user's face, unlocking the door and providing a voice prompt that the door is unlocked. This invention improves the accuracy of smart door lock recognition without affecting normal user operation, reduces false unlocking, and ensures that the door is only unlocked when the user intends to and meets specific security conditions, thus improving security.

[0100] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or smart lock system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or smart lock system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or smart lock system that includes that element.

[0101] Of course, those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.). The program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The computer-readable storage medium can be a memory, magnetic disk, optical disk, etc.

[0102] It should be understood that the application of the present invention is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A face recognition method for smart door locks based on attitude angle, characterized in that, The smart door lock face recognition method based on attitude angle includes: When a face is detected, the current user's face image and the current time are obtained. Based on the historical door opening and closing record data, the historical time of the most recent time when the user closed the door lock is obtained. The time interval is calculated based on the current time and the historical time. Determine whether the time interval is less than a preset time threshold. If the time interval is less than the preset time threshold, confirm the face posture based on the current user's face image to obtain the flip angle, pitch angle, and yaw angle. Determine whether the roll angle, pitch angle and yaw angle simultaneously meet the preset angle range conditions. If they simultaneously meet the preset angle range conditions, confirm whether the current user's face image is a registered user's face. If it is a registered user's face, unlock the door and give a voice prompt that the door lock is unlocked. The step of determining whether the time interval is less than a preset time threshold, and if the time interval is less than the preset time threshold, confirming the facial posture based on the current user's facial image to obtain the flip angle, pitch angle, and yaw angle, specifically includes: A preset time threshold is set in advance, and it is determined whether the time interval is less than the preset time threshold. If the time interval is less than the preset time threshold, the face region in the face image is determined using a face detection algorithm; The key feature points of the face are located using a feature point detection algorithm, and the roll angle, pitch angle and yaw angle in the attitude parameters are determined based on the key feature points. The process involves determining whether the roll angle, pitch angle, and yaw angle simultaneously meet a preset angle range condition. If they do, it is confirmed whether the current user's face image is that of a registered user. If it is, the door lock is opened, and a voice prompt indicates that the door lock is open. Specifically, this includes: A preset angle range condition is set in advance, wherein the angle change range is within a preset angle, and it is determined whether the change ranges of the roll angle, the pitch angle and the yaw angle are all within the preset angle; If the range of change of the flip angle, the pitch angle and the yaw angle are all within the preset angle, it indicates that the current user is facing the door lock. Then it is determined whether the face image of the current user is a registered user's face. If the current user's face image is identified as that of a registered user, the door lock is opened, and a voice prompt is given to the current user that the door lock is open. By limiting the recognition angle of the facial recognition module and introducing a time threshold, the system effectively prevents users from accidentally triggering the facial unlocking function after opening the door due to turning around or adjusting their posture, thereby improving the security and reliability of the smart door lock.

2. The face recognition method for smart door locks based on attitude angle according to claim 1, characterized in that, When a face is detected, the current user's facial image and the current time are acquired. Based on historical door opening and closing records, the most recent time the user closed the door is obtained. A time interval is calculated based on the current time and the historical time. Specifically, this includes: Acquire the current user's facial image captured by the depth camera, and obtain the current time; The face image is preprocessed, including noise reduction, contrast enhancement, color correction, and image normalization. Obtain historical door opening and closing records, and based on the historical door opening and closing records, obtain the historical time of the most recent time when the user closed the door lock. Calculate the time interval based on the current time and the historical time.

3. The face recognition method for smart door locks based on attitude angle according to claim 1, characterized in that, The step of determining whether the time interval is less than a preset time threshold further includes: If the time interval is greater than or equal to the preset time threshold, then it is determined whether the current user's face image is a registered user's face; If the current user's facial image is that of a registered user, then unlock the door. If the current user's face image is not a registered user's face, the door lock will not be opened.

4. The face recognition method for smart door locks based on attitude angle according to claim 1, characterized in that, The step of determining whether the ranges of change of the roll angle, the pitch angle, and the yaw angle are all within a preset angle includes: If the range of change of any one of the roll angle, pitch angle and yaw angle exceeds the preset angle, it indicates that the current user is not facing the door lock, and the door lock will not be opened.

5. The face recognition method for smart door locks based on attitude angle according to claim 1, characterized in that, The step of determining whether the current user's facial image is that of a registered user also includes: If the current user's face image is not identified as belonging to a registered user, the door lock will not be opened.

6. A smart door lock face recognition system based on attitude angle, wherein the smart door lock face recognition system based on attitude angle is used to implement the smart door lock face recognition method based on attitude angle according to any one of claims 1-5, characterized in that, The smart door lock face recognition system based on attitude angle includes: The face recognition module is used to acquire the current user's face image and current time when a face is detected, acquire the historical time of the most recent time the user closed the door lock based on historical door opening and closing records, and calculate the time interval based on the current time and the historical time. The posture detection module is used to determine whether the time interval is less than a preset time threshold. If the time interval is less than the preset time threshold, the face posture is confirmed based on the current user's face image to obtain the flip angle, pitch angle and yaw angle. The door lock control module is used to determine whether the roll angle, pitch angle and yaw angle simultaneously meet the preset angle range conditions. If they simultaneously meet the preset angle range conditions, it confirms whether the current user's face image is a registered user's face. If it is a registered user's face, the door lock is opened and a voice prompt is given that the door lock is open.

7. A smart door lock, characterized in that, The smart lock includes: a memory, a processor, and a smart lock face recognition program based on attitude angle stored in the memory and executable on the processor. When the smart lock face recognition program based on attitude angle is executed by the processor, it implements the steps of the smart lock face recognition method based on attitude angle as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a smart door lock face recognition program based on attitude angles, which, when executed by a processor, implements the steps of the smart door lock face recognition method based on attitude angles as described in any one of claims 1-5.

Images