Identity verification method of intelligent cabinet, intelligent cabinet and storage medium

Abstract

The invention discloses an identity verification method of an intelligent cabinet, the intelligent cabinet and a storage medium, and relates to the technical field of intelligent cabinets, and the method comprises the steps: obtaining facial biological feature data and real-time movement data of a user; inputting the facial biological feature data and a pre-stored user identity library into a pre-trained identity verification model, so that the identity verification model outputs an identity verification result; wherein all authorized facial feature data are stored in the user identity library; according to the identity verification result and the real-time mobile data, generating a temporary operation authority with duration constraint; and carrying out required parcel access operation based on the temporary operation authority. According to the scheme, the problem that an existing intelligent cabinet depends on a single password unlocking mechanism, so that the use safety is low is solved, and the use safety of the intelligent cabinet is effectively improved.

Description

Authentication methods for smart lockers, smart lockers and storage media Technical Field

[0001] This invention relates to the field of smart cabinet technology, and in particular to a smart cabinet authentication method, a smart cabinet, and a storage medium. Background Technology

[0002] Smart lockers, as a crucial infrastructure for last-mile delivery, have been widely deployed in communities, industrial parks, and public service settings. Their basic function is to provide users with self-service parcel storage and retrieval. However, current smart locker authentication methods primarily rely on one-time SMS passwords or fixed passwords for unlocking. Such single-password mechanisms pose significant security risks in practical use.

[0003] For example, during the user package pickup process, passwords can easily be spied on, intercepted, or forwarded by others during transmission or input, allowing unauthorized individuals to complete the pickup operation by obtaining the password. During the courier delivery process, existing smart lockers typically use fixed employee IDs or simple passwords for verification, posing a risk of identity theft and failing to effectively verify the delivery person's true identity. In other words, existing smart lockers suffer from insufficient identity verification at both the user pickup and courier delivery stages, making package security dependent on the confidentiality of the password itself.

[0004] Therefore, improving the security of smart cabinets has become a pressing technical problem that needs to be solved in this field. Summary of the Invention

[0005] This invention provides an authentication method for a smart cabinet, a smart cabinet, and a storage medium to solve the problem that existing smart cabinets rely on a single password unlocking mechanism, resulting in low security.

[0006] In a first aspect, a method for authenticating a smart locker is provided, comprising: acquiring a user's facial biometric data and real-time movement data; inputting the facial biometric data and a pre-stored user identity database into a pre-trained authentication model, so that the authentication model outputs an identity verification result; wherein, the user identity database stores all authorized facial feature data; generating temporary operation permissions with duration constraints based on the identity verification result and the real-time movement data; and performing the required package storage and retrieval operations based on the temporary operation permissions.

[0007] Secondly, a smart locker is provided, comprising a control module; the control module is used to: acquire a user's facial biometric data and real-time movement data; input the facial biometric data and a pre-stored user identity database into a pre-trained identity verification model, so that the identity verification model outputs an identity verification result; wherein, the user identity database stores all authorized facial feature data; generate temporary operation permissions with duration constraints based on the identity verification result and the real-time movement data; and perform the required package storage and retrieval operations based on the temporary operation permissions.

[0008] Thirdly, a smart cabinet is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the aforementioned smart cabinet authentication method.

[0009] Fourthly, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described intelligent cabinet authentication method.

[0010] This invention provides a technical solution comprising: acquiring a user's facial biometric data and real-time movement data; inputting the facial biometric data and a pre-stored user identity database into a pre-trained identity verification model, so that the identity verification model outputs an identity verification result; wherein, the user identity database stores all authorized facial feature data; generating temporary operation permissions with duration constraints based on the identity verification result and the real-time movement data; and performing the required package storage and retrieval operations based on the temporary operation permissions. This solution improves the accuracy and anti-interference capability of identity verification by sequentially comparing facial biometric data with all authorized facial feature data in the user identity database through a pre-trained identity verification model; it effectively prevents permission abuse by generating temporary operation permissions with duration constraints based on the identity verification result and real-time movement data; and finally, it enables users to perform the required package storage and retrieval operations through temporary operation permissions, effectively reducing the identity misidentification rate and the risk of impersonation, and improving the security of smart lockers. Attached Figure Description

[0011] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For users of ordinary skills in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 is a schematic diagram of a smart cabinet according to an embodiment of the present invention; Figure 2 is a flowchart of an authentication method for a smart cabinet according to an embodiment of the present invention; Figure 3 is another flowchart of an authentication method for a smart cabinet according to an embodiment of the present invention; Figure 4 is another flowchart of an authentication method for a smart cabinet according to an embodiment of the present invention; Figure 5 is another flowchart of an authentication method for a smart cabinet according to an embodiment of the present invention; Figure 6 is another flowchart of an authentication method for a smart cabinet according to an embodiment of the present invention; Figure 7 is another flowchart of an authentication method for a smart cabinet according to an embodiment of the present invention; Figure 8 is another flowchart of an authentication method for a smart cabinet according to an embodiment of the present invention; Figure 9 is another schematic diagram of a smart cabinet according to an embodiment of the present invention. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by users of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0014] The present invention provides an authentication method for smart lockers, which can be applied to smart lockers. The smart locker may include parcel storage devices with automatic identification, network communication, and remote control capabilities, such as smart express lockers, smart fresh food lockers, smart home lockers, smart office document exchange lockers, smart shared resource requisition lockers, and smart vending machines, etc., without limitation herein.

[0015] The smart cabinet typically consists of multiple independently controllable storage compartments (slots / boxes), and its hardware system architecture may include, but is not limited to, the components shown in Figure 1: a control module 101 (e.g., a central processing unit (CPU), microcontroller unit (MCU), or dedicated controller (such as a field-programmable gate array (FPGA), digital signal processor (DSP), etc., which are not limited here), an image acquisition device 102, a millimeter-wave radar 103, an identification module 104 (e.g., a Bluetooth beacon identification unit, a UWB identity binding module, etc.), an execution module 105 (e.g., an electronic lock), a voice module 106, and a communication module 107 (e.g., a 4G module, a 5G module, a Wi-Fi module, etc.). The various hardware modules interact in real time via a system bus or other data transmission methods, and through collaborative work, they jointly execute the steps of the smart cabinet's authentication method.

[0016] Please refer to Figures 2 through 8 below for a detailed explanation of the authentication method for a smart cabinet provided by an embodiment of the present invention. It should be noted that, for ease of explanation of the embodiments of the present invention, the relevant hardware components of the smart cabinet will not be identified in the subsequent method description.

[0017] Example A, as shown in Figure 2, provides an authentication method for a smart cabinet. Taking the application of this method in the smart cabinet shown in Figure 1 as an example, the method includes the following steps: S201: Obtain the user's facial biometric data and real-time movement data.

[0018] As an example, real-time facial images of the user can be captured using an image acquisition device (e.g., a camera), and these images can be directly used as facial biometric data. Alternatively, face detection, key point localization, and pose correction can be performed on the captured real-time facial images to extract real-time facial feature vectors, which can then be used as facial biometric data; this is not limited here. Simultaneously, real-time movement data of the user can be collected using millimeter-wave radar. This real-time movement data is a set of dynamic behavioral parameters of the user within the sensing area of ​​the smart cabinet, specifically including but not limited to parameters such as the real-time distance between the user and the cabinet, movement speed, movement trajectory, and motion posture characteristics; this is not limited here.

[0019] S202: Input facial biometric data and a pre-stored user identity database into a pre-trained identity verification model so that the identity verification model outputs an identity verification result.

[0020] In this embodiment, the pre-trained authentication model can be a facial recognition model trained based on a deep learning architecture (such as ArcFace or CosFace), and there is no limitation here.

[0021] As an example, the control module uses the real-time facial image or real-time facial feature vector obtained in step S201 as input data, and simultaneously calls a pre-established and maintained user identity database. The storage location of this user identity database includes the local storage unit of the smart cabinet or a cloud server, where authorized facial feature data of all registered and authorized users is stored. The format of this authorized facial feature data is strictly consistent with the format of the real-time acquired facial biometric data: when the real-time facial biometric data is a real-time facial image, the user identity database stores the authorized facial image; when the real-time facial biometric data is a real-time facial feature vector, the user identity database stores the authorized facial feature vector.

[0022] The control module inputs facial biometric data along with all authorized facial feature data from the user identity database into a pre-trained authentication model. This authentication model is trained based on a deep learning architecture and processes data through a specific algorithm: if the input is a facial image, the authentication model extracts features and calculates similarity between the real-time facial image and each authorized facial image in the user identity database; if the input is a facial feature vector, the authentication model directly calculates the similarity between the real-time facial feature vector and each authorized facial feature vector in the user identity database.

[0023] Finally, the identity verification model outputs the identity verification result based on the calculated similarity. Specifically, a preset similarity threshold is used as the judgment standard during the process. When the highest similarity exceeds this threshold, the identity verification model outputs a verified result; when the highest similarity does not reach the threshold, the identity verification model outputs a failed identity verification result. The similarity threshold can range from 0.65 to 0.98, preferably 0.75, 0.85, or 0.90, and is not limited here. The identity verification result can include verified success, verified failure, and the corresponding identity information.

[0024] S203: Generate temporary operation permissions with duration constraints based on identity verification results and real-time mobile data.

[0025] In this embodiment, the duration constraint can be the validity period of a pre-selected temporary operation permission, such as 5 minutes, 10 minutes, or 15 minutes, and is not limited here.

[0026] As an example, based on the identity verification result and real-time movement data, temporary operation permissions with duration constraints are generated. That is, when the identity verification result is successful and the real-time movement data indicates that the user intends to approach the cabinet (for example, the relative distance to the cabinet is less than a threshold and the approach speed or direction clearly indicates an intention to approach the cabinet), temporary operation permissions with duration constraints are generated. In this way, the real-time granting and revoke of temporary delivery permissions based on identity verification results and approach intentions is realized, which not only ensures the accuracy of authorization timing, but also effectively prevents the risk of permission abuse by setting a validity period.

[0027] S204: Based on temporary operation permissions, perform the required package storage and retrieval operations.

[0028] As an example, based on temporary operation permissions, the required package storage and retrieval operations can be performed. That is, two parallel operation paths can be provided. For instance, within the validity period of the temporary operation permissions, the recipient can choose to retrieve the package using a traditional retrieval code; or directly retrieve the associated package based on the recipient's identifier in the identity verification result and control the corresponding locker compartment to open and complete the retrieval. Similarly, the courier can also complete the delivery by verifying their identity using traditional verification information (such as dynamic passwords or employee ID verification); or directly query the package information of the package associated with the courier's identifier based on the courier's identifier in the identity verification result to complete the delivery. It is evident that the identity verification method of the smart locker provided by this invention ensures high security regardless of whether it is a traditional verification method or a novel contactless verification method.

[0029] In summary, the technical solution provided by this invention includes: acquiring a user's facial biometric data and real-time movement data; inputting the facial biometric data and a pre-stored user identity database into a pre-trained identity verification model, so that the identity verification model outputs an identity verification result; wherein, the user identity database stores all authorized facial feature data; generating temporary operation permissions with duration constraints based on the identity verification result and the real-time movement data; and performing the required package storage and retrieval operations based on the temporary operation permissions. This solution improves the accuracy and anti-interference capability of identity verification by sequentially comparing facial biometric data with all authorized facial feature data in the user identity database through a pre-trained identity verification model; it effectively prevents permission abuse by generating temporary operation permissions with duration constraints based on the identity verification result and real-time movement data; and finally, it enables users to perform the required package storage and retrieval operations through temporary operation permissions, effectively reducing the identity misidentification rate and the risk of impersonation, and improving the security of smart lockers.

[0030] In one embodiment, as shown in FIG3, step S201, which involves acquiring the user's facial biometric data and real-time movement data, includes the following steps: S211: Real-time detection of whether the user has entered the sensing area; S212: When the user is detected to have entered the sensing area, the user's facial biometric data is acquired through an image acquisition device, and the user's real-time movement trajectory and the real-time distance between the user and the smart cabinet are acquired through millimeter-wave radar; S213: The real-time movement trajectory and the real-time distance are used as real-time movement data.

[0031] In this embodiment, the sensing area is a pre-defined data collection area, such as a rectangular area formed by extending 2.5 meters forward and 1 meter to the left and right sides from the center of the front of the smart cabinet as a reference point. This is only an example and does not constitute a limitation of the present invention.

[0032] As an example, when a user is detected entering the sensing area, the image acquisition device can be controlled to collect the user's facial biometric data, and simultaneously the millimeter-wave radar can be controlled to continuously scan the user entering the sensing area to obtain their real-time movement trajectory around the cabinet, and the real-time distance between the user and the smart cabinet can be calculated; the real-time movement trajectory and the real-time distance are combined to form real-time movement data. It should be understood that the above-mentioned real-time movement data may also include parameters such as movement speed and motion posture characteristics, which are not limited here.

[0033] For example, when a user is detected entering the sensing area, the image acquisition device and the millimeter-wave radar work in parallel. Specifically, the image acquisition device can capture real-time images of the user entering the sensing area at a frame rate of 30 fps, performing face detection, key point localization, and posture correction. When the target face reaches a recognizable distance (e.g., 0.5-2.5 meters), it captures a clear facial image and / or extracts facial feature vectors to obtain facial biometric data. Simultaneously, the millimeter-wave radar continuously scans the user entering the sensing area at an update frequency of 20 Hz, outputting time-series location points and calculating the real-time movement trajectory (e.g., a time-ordered displacement vector sequence (x(t), y(t))) and the real-time distance from the cabinet reference point (calculated using Euclidean distance). The trajectory and distance are synchronized with timestamps and packaged into real-time movement data (e.g., in a structured record format of <timestamp, position, distance, speed, direction>). In this way, both facial samples that meet the recognition requirements and real-time movement data confirming the user's intention to approach are obtained.

[0034] In one embodiment, as shown in Figure 4, step 103, which is generating temporary operation permissions with duration constraints based on identity verification results and real-time mobile data, includes the following steps: S231: Outputting real-time mobile data to a pre-trained intent recognition model in real time, so that the intent recognition model can analyze the real-time mobile data in real time and output intent recognition results; S232: Generating temporary operation permissions with duration constraints based on identity verification results and intent recognition results.

[0035] In this embodiment, the pre-trained intent recognition model can be an intent recognition model trained based on a deep learning architecture (such as ArcFace or CosFace), and there is no limitation on this. It should be understood that the above-mentioned identity verification model and intent recognition model can be encapsulated into an AI assistant, that is, identity verification and intent recognition can be achieved through an AI assistant, and there is no limitation on this.

[0036] As an example, the real-time mobile data obtained in step S101 is transmitted in real-time to a pre-trained intent recognition model. The intent recognition model performs online analysis on the real-time mobile data, including real-time movement trajectory and real-time distance, and outputs intent recognition results. These results can include either an intent to approach the cabinet or no intent to approach the cabinet. Next, if the identity verification result is successful and the intent recognition result indicates an intent to approach the cabinet, a temporary operation permission with duration constraints is generated. If the identity verification result is unsuccessful but the intent recognition result indicates an intent to approach the cabinet, a temporary operation permission with duration constraints can be generated based on the first weight corresponding to the identity verification result and the second weight corresponding to the intent recognition result. For example, a comprehensive evaluation can be performed based on a preset weight allocation scheme (the first weight of the identity verification result is 0.7, and the second weight of the intent recognition result is 0.3). If the weighted score exceeds a preset threshold, a temporary operation permission with duration constraints is generated; otherwise, authorization is denied. This effectively reduces the identity misidentification rate and the risk of impersonation.

[0037] In one embodiment, as shown in FIG5, step S232, which is the generation of temporary operation permissions with time constraints based on identity verification results and intent recognition results, includes the following steps: S2321: Obtain the device identification information of the smart terminal held by the user; S2322: Match the device identification information with the authorized device identification information in the target authorization list to obtain the matching result; S2323: Generate temporary operation permissions with time constraints based on identity verification results, intent recognition results and matching results.

[0038] As an example, a smart terminal refers to a communication device that is close to a user and has a unique device identifier. It may include, but is not limited to, smartphones, wearable devices (e.g., smartwatches) or handheld terminals (e.g., PDA devices (personal digital assistants, which are portable smart devices), etc., without limitation.

[0039] Device identification information is coded information that can uniquely identify the smart terminal, such as Bluetooth MAC address, RFID serial number, UWB tag number or dynamic encryption token, etc., and is not limited here.

[0040] The target authorization list refers to a pre-configured authorization information database used to record the identification information of authorized devices permitted to operate the smart cabinet. This target authorization list can be stored in the smart cabinet's local database or in the cloud, and can be updated in real time according to usage, which does not constitute a limitation of this invention.

[0041] As an example, the identification module can automatically collect device identification information near the user's smart terminal and match it item by item with the authorized device identification information in the target authorization list synchronized in the cloud or locally to obtain the matching result. When the same device identification information is found in the target authorization list, a matching success result (i.e., device authentication passed) is obtained; when no matching authorized device identification information is found, a matching failure result (i.e., device authentication failed) is obtained. In the specific matching process, string matching algorithms, hash-based fast matching algorithms, or other algorithm types suitable for efficient retrieval can be used, without limitation.

[0042] Next, based on the identity verification result, intent recognition result, and matching result, a temporary operation permission with duration constraints is generated. Specifically, when the identity verification result is successful, the intent recognition result indicates an intent to approach the cabinet, and the matching result is a successful match, a temporary operation permission with duration constraints is generated. Similarly, when the identity verification result is unsuccessful, the intent recognition result indicates no intent to approach the cabinet, and the matching result is a failed match, a temporary operation permission with duration constraints can be generated based on the first weight corresponding to the identity verification result, the second weight corresponding to the intent recognition result, and the third weight corresponding to the matching result. For example, a comprehensive evaluation can be performed based on a preset weight allocation scheme (the first weight of the identity verification result is 0.5, the second weight of the intent recognition result is 0.3, and the third weight of the matching result is 0.2). If the weighted score exceeds a preset threshold, a temporary operation permission with duration constraints is generated; otherwise, authorization is denied. In this way, the identity misidentification rate and the risk of impersonation can be further effectively reduced.

[0043] In one embodiment, as shown in Figure 6, step S204, which is the necessary package storage and retrieval operation based on temporary operation permissions, includes the following steps: S241A: When the user's identity type is a pickup person, obtain the pickup person's identifier corresponding to the pickup person; S242A: Query the package to be picked up associated with the pickup person's identifier and determine the first target compartment where the package to be picked up is located; S243A: Based on temporary operation permissions, control the door of the first target compartment to open to complete the retrieval.

[0044] As an example, when a user's identity type is "pickup person," the identity information is first parsed based on the identity verification result to obtain the pickup person's identifier. Then, using this pickup person identifier as the search key, all associated pending package information is queried, and the first target locker corresponding to the pending package is determined. After successfully locating the first target locker, the execution module, based on the temporary operation permission generated in step S103, controls the opening of the locker door, allowing the pickup person to complete the pickup operation from the corresponding first target locker. After pickup, the locker door automatically closes, and upon detecting that the door is closed, the temporary operation permission is immediately invalidated, thus ending the pickup process. Through this method, seamless pickup without SMS verification codes or QR code scanning can be achieved, effectively eliminating the cumbersome steps caused by manual input in traditional pickup processes and significantly optimizing the user experience.

[0045] Taking a typical user package retrieval scenario at a smart locker as an example, after user A (identity type: retrieval person) completes identity verification, the system automatically obtains their retrieval person identifier (e.g., user ID: C2x2x0111). Then, based on this identifier, the system queries the backend database, matching it with a linked package (tracking number: YTxxxxxx111), and determines that the package is located in the first target compartment (medium-sized compartment, 35cm×25cm×20cm) on the third floor of the smart locker. During the period when user A's temporary access permission (e.g., valid for 5 minutes) is active, an unlock command is sent to the execution module of the first target compartment. This allows the execution module to control the automatic opening of the door of compartment 5 on the third floor, while the voice module simultaneously plays a prompt: "You have a new package, please retrieve it." After user A retrieves the package, the door automatically closes, the temporary access permission expires immediately, and the retrieval process is complete. This effectively avoids the risk of remote fraudulent retrieval using a leaked password, as is common in traditional solutions, while also reducing the manual steps required for the retrieval process, further improving the convenience and security of the smart locker.

[0046] In one embodiment, as shown in Figure 7, step S204, which involves performing the required package storage and retrieval operation based on temporary operation permissions, includes the following steps: S241B: When the user's identity type is a courier, obtain the courier identifier corresponding to the courier; S242B: Query the package information of the package to be delivered associated with the courier identifier, including package size information; S243B: Obtain the compartment information of the smart locker, including compartment size information; S244B: Determine the second target compartment based on the package size information and compartment size information; S245B: Based on temporary operation permissions, control the door corresponding to the second target compartment to open to complete the delivery.

[0047] As an example, when the user's identity type is a courier, the system first parses the identity information based on the identity verification result to obtain the courier's identifier. Then, it retrieves the package information associated with the courier identifier and extracts necessary parameters such as package size information. Simultaneously, it obtains the compartment information for each type of smart locker and extracts the compartment size information for each type. Using a preset intelligent matching algorithm (e.g., based on volume-priority matching, size threshold judgment, three-dimensional adaptation of length, width, and height, or size range mapping), it compares and analyzes the package size information with the compartment size information to determine the most suitable second target compartment for this delivery. After successfully locating the second target compartment, the execution module, based on the temporary operation permission generated in step S103, controls the automatic opening of the locker door corresponding to the second target compartment and simultaneously plays a prompt message such as "Please deliver the package" via the voice module. After the courier places the package into the second target compartment, the locker door automatically closes, and the temporary operation permission is revoked after confirming the door is closed.

[0048] As can be seen, this embodiment uses the courier's identifier as the retrieval entry point, structurally associating the information of the package to be delivered with the information of the locker compartment. It then uses an intelligent matching algorithm between the package size and the locker compartment size to accurately select the second target locker compartment. This effectively solves the problem of smart lockers generally using fixed compartment designs, where the compartment size and quantity are fixed at the factory, making it impossible to dynamically adjust space allocation according to package volume. This often leads to resource waste ("using a large locker for a small package") or "a small locker not being able to hold a large package" in practical applications. This embodiment ensures the size compatibility of packages placed in the locker compartments (avoiding "a small locker not being able to hold a package") and balances the usage load of different compartments, improving the utilization rate of locker space, reducing resource waste, and lowering the probability of delivery failure due to fixed compartment sizes. This, in turn, improves the reliability of package delivery and the overall operational efficiency of the smart locker.

[0049] In one embodiment, as shown in FIG8, step S244B, which involves determining the second target box based on the package size information and the box size information, includes the following steps: S244B1: Determine the first box type based on the package size information and the box size information; S244B2: Determine whether there is a first available box in the first box type; S244B3: If there is, select one first available box from all first available boxes of the first box type as the second target box using a preset random algorithm; S244B4: If there is no first available box, determine the second box type associated with the first box type, and select one second available box from all second available boxes of the second box type as the second target box using a preset random algorithm.

[0050] As an example, when the user's identity type is a courier, the system first obtains the package size information of the parcel to be delivered, and then analyzes the size compatibility between the parcel and the parcel in conjunction with the size information of various types of lockers in the smart locker to determine the first locker type suitable for the current parcel. For example, when the parcel size is 30 cm × 20 cm × 15 cm, it can be identified as a parcel suitable for the "medium-sized locker", thus determining the medium-sized locker as the first locker type.

[0051] Subsequently, the availability status of all medium-sized lockers is checked to determine if there is an available first-choice locker. For example, if medium-sized lockers numbered 3-5, 3-7, and 4-2 are all found to be idle, it is determined that there is an available first-choice locker. In this case, a preset random algorithm is invoked to randomly select one of the available medium-sized lockers as the second-choice locker. For example, the preset random algorithm selects the medium-sized locker number 7 on the 3rd floor as the final second-choice locker. By randomly selecting, a balanced distribution of delivery load among lockers of the same type can be achieved while ensuring size compatibility, effectively avoiding long-term high loads on lockers caused by fixed-order deliveries.

[0052] If all medium-sized compartments are determined to be occupied, proceed to step S244B4 to determine the second compartment type associated with the medium-sized compartment type. For example, the associated type of "medium-sized compartment" is "large-sized compartment". After identifying all available large-sized compartments, a preset random algorithm is still used to randomly select one of the available large-sized compartments as the second target compartment to ensure that the package can still be successfully delivered.

[0053] It should be understood that when there is no available first container for the first container type and no available second container for the associated second container type, the current delivery operation can be paused; or, the third container type associated with the second container type can be determined, and the same availability judgment and random selection process can be executed in sequence until there are no available containers before the current delivery is paused, thereby maximizing the delivery success rate.

[0054] In summary, this embodiment first determines the preferred locker type based on package type and size information, then performs availability checks on that type, and selects a specific locker using a preset random algorithm when available. If unavailable, it automatically reverts to other associated locker types for the same random selection, achieving dynamic and precise selection of the second target locker. This strategy ensures compatibility between packages and lockers in terms of size and function, effectively avoiding direct mismatches such as "lockers being too small to accommodate packages," and effectively distributes delivery load through randomized selection, reducing overuse of popular lockers. The associated fallback mechanism when the preferred locker type is unavailable prevents delivery interruptions due to the exhaustion of a particular type of locker, improving delivery success rate and overall locker availability. Furthermore, binding a selected locker to temporary operation permissions and immediately revoking the permissions after delivery strictly limits opening permissions to the target locker and the valid time period, reducing the risk of misdelivery and unauthorized opening. Overall, this embodiment improves the utilization rate of cabinet space, delivery adaptability and scheduling flexibility, while further enhancing the security and controllability of the smart cabinet delivery process.

[0055] In one embodiment, namely in step S244B1, the package information further includes package category information, and the compartment information further includes compartment storage category information; that is, determining the first compartment type based on the package size information and the compartment size information includes: S244B11: determining the first compartment type based on the package size information, compartment size information, package category information and compartment storage category information.

[0056] As an example, when a courier delivers a package, they not only obtain the package's dimensions such as length, width, and height, but also extract the package's category information, such as "fresh produce," "medicine," and "room temperature goods." Simultaneously, the smart locker's compartment information is pre-configured with the category information for each compartment; for example, some compartments are labeled "supports refrigerated items (2–8℃)," "can store medicine," and "room temperature compartment."

[0057] When step S244B11 is executed, a basic size matching judgment is first performed using the package size information and the compartment size information to filter out a set of compartment types that can accommodate the package. Then, a second filtering is performed by combining the package category information and the compartment storage category information to ensure that the selected compartments not only match the size but also meet the package category requirements in terms of storage attributes. For example, if the package size matches a medium-sized compartment, but the package category is "fresh produce," all ordinary medium-sized compartments will be excluded, and only the "fresh produce-compatible medium-sized compartment" with temperature control will be retained as a candidate compartment type. Thus, the first compartment type is finally determined, for example, "fresh produce medium-sized compartment."

[0058] In subsequent steps S244B2-S244B4, it will be further determined whether there is a first available container in the first container type based on the determined first container type; if there is, one of them will be selected as the second target container; if there is no, it will be further switched to the associated second container type to ensure that the delivery container meets the requirements in both size and category compatibility.

[0059] As can be seen, this embodiment can intelligently select the first compartment type that can both accommodate the package and meet the storage attribute requirements based on the package size information and package category information. It achieves dual matching of size adaptation and category adaptation, which not only ensures the size adaptability of the package into the compartment (avoiding "the compartment is too small to hold the package"), but also meets the diverse storage needs of items, such as the storage requirements of special categories such as fresh food and medicine, thus improving the user experience.

[0060] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0061] In a second aspect of embodiment B, a smart locker is provided, as shown in Figure 1. The smart locker includes a control module 101. The control module 101 is configured to: acquire a user's facial biometric data and real-time movement data; input the facial biometric data and a pre-stored user identity database into a pre-trained identity verification model, so that the identity verification model outputs an identity verification result; wherein the user identity database stores all authorized facial feature data; generate temporary operation permissions with duration constraints based on the identity verification result and the real-time movement data; and perform the required package storage and retrieval operations based on the temporary operation permissions.

[0062] In one embodiment, the smart cabinet further includes an image acquisition device 102 and a millimeter-wave radar 103; the control module 101 is further configured to: detect in real time whether the user enters the sensing area; when the user is detected to have entered the sensing area, acquire the facial biometric data of the user acquired by the image acquisition device 102, and the real-time movement trajectory of the user acquired by the millimeter-wave radar 103, as well as the real-time distance between the user and the smart cabinet; and use the real-time movement trajectory and the real-time distance as the real-time movement data.

[0063] In one embodiment, the control module 101 is further configured to: output the real-time mobile data to a pre-trained intent recognition model in real time, so that the intent recognition model can perform real-time analysis on the real-time mobile data and output intent recognition results; and generate temporary operation permissions with duration constraints based on the identity verification results and the intent recognition results.

[0064] In one embodiment, the control module 101 is further configured to: obtain device identification information of the smart terminal held by the user; match the device identification information with authorized device identification information in the target authorization list to obtain a matching result; and generate temporary operation permissions with duration constraints based on the identity verification result, the intent recognition result and the matching result.

[0065] In one embodiment, the control module 101 is further configured to: when the user's identity type is a pickup person, obtain the pickup person identifier corresponding to the pickup person; query the package to be picked up associated with the pickup person identifier, and determine the first target compartment where the package to be picked up is located; and based on the temporary operation permission, control the door corresponding to the first target compartment to open to complete the pickup.

[0066] In one embodiment, the control module 101 is further configured to: when the user's identity type is a courier, obtain the courier identifier corresponding to the courier; query the package information of the package to be delivered associated with the courier identifier, the package information including package size information; obtain the compartment information of the types of compartments provided in the smart cabinet, the compartment information including compartment size information; determine the second target compartment based on the package size information and the compartment size information; and control the door corresponding to the second target compartment to open to complete the delivery based on the temporary operation permission.

[0067] In one embodiment, the control module 101 is further configured to: determine a first compartment type based on the package size information and the compartment size information; determine whether a first available compartment exists in the first compartment type; if it exists, select a first available compartment from all first available compartments of the first compartment type as the second target compartment using a preset random algorithm; if it does not exist, determine a second compartment type associated with the first compartment type, and select a second available compartment from all second available compartments of the second compartment type as the second target compartment using the preset random algorithm.

[0068] For specific limitations regarding smart lockers, please refer to the limitations on the authentication methods for smart lockers mentioned above, which will not be repeated here.

[0069] In a third aspect of embodiment C, a smart cabinet is provided. In one embodiment, a computer device is provided, the internal structure of which can be shown in Figure 9. The computer device includes a processor, a memory, and a network interface connected via a system bus. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. When the computer program is executed by the processor, it implements the steps of an authentication method for the aforementioned smart cabinet.

[0070] In one embodiment, a smart cabinet is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the authentication method of the smart cabinet in the above embodiment, such as S201-S204 shown in FIG2, or as shown in FIG2 to FIG8. To avoid repetition, these steps will not be described again here. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in this embodiment of the smart cabinet, such as the function of the authentication method of the smart cabinet shown in FIG1. ​​To avoid repetition, these steps will not be described again here.

[0071] In a fourth aspect of Embodiment D, a computer-readable storage medium is provided, on which a computer program is stored. When executed by a processor, the computer program implements the steps of the authentication method for the smart cabinet in the above embodiments, such as S201-S204 shown in FIG2, or as shown in FIG2 to FIG8. To avoid repetition, these steps will not be repeated here. Alternatively, when executed by a processor, the computer program implements the functions of each module / unit in this embodiment of the smart cabinet, such as the smart cabinet control function shown in FIG1. ​​To avoid repetition, these steps will not be repeated here. The computer-readable storage medium can be non-volatile or volatile.

[0072] Users skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0073] Those skilled in the art will understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0074] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, users of ordinary skills in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for identity verification of a smart locker, characterized in that, include: Acquire the user's facial biometric data and real-time movement data; input the facial biometric data and a pre-stored user identity database into a pre-trained identity verification model, so that the identity verification model outputs an identity verification result; wherein, the user identity database stores all authorized facial feature data; generate temporary operation permissions with duration constraints based on the identity verification result and the real-time movement data; Based on the aforementioned temporary operation permissions, perform the required package storage and retrieval operations.

2. The authentication method for the smart cabinet as described in claim 1, characterized in that, The acquisition of the user's facial biometric data and real-time movement data includes: real-time detection of whether the user enters the sensing area; when the user enters the sensing area, acquiring the user's facial biometric data through an image acquisition device, and acquiring the user's real-time movement trajectory and the real-time distance between the user and the smart cabinet through millimeter-wave radar; and using the real-time movement trajectory and the real-time distance as the real-time movement data.

3. The authentication method for the smart cabinet as described in claim 1, characterized in that, The step of generating temporary operation permissions with duration constraints based on the identity verification result and the real-time mobile data includes: outputting the real-time mobile data to a pre-trained intent recognition model in real time, so that the intent recognition model can analyze the real-time mobile data in real time and output intent recognition results; and generating temporary operation permissions with duration constraints based on the identity verification result and the intent recognition results.

4. The authentication method for the smart cabinet as described in claim 3, characterized in that, The step of generating a temporary operation permission with duration constraints based on the identity verification result and the intent recognition result includes: obtaining the device identification information of the smart terminal held by the user; matching the device identification information with the authorized device identification information in the target authorization list to obtain a matching result; and generating a temporary operation permission with duration constraints based on the identity verification result, the intent recognition result, and the matching result.

5. The authentication method for the smart cabinet as described in claim 1, characterized in that, The step of performing the required package storage and retrieval operations based on the temporary operation permission includes: when the user's identity type is a pickup person, obtaining the pickup person's identifier; querying the package to be picked up associated with the pickup person's identifier and determining the first target compartment where the package to be picked up is located; and controlling the door of the first target compartment to open to complete the pickup based on the temporary operation permission.

6. The authentication method for the smart cabinet as described in claim 1, characterized in that, The process of performing the required package storage and retrieval operations based on the temporary operation permission includes: when the user's identity type is a courier, obtaining the courier's corresponding courier identifier; querying the package information of the package to be delivered associated with the courier identifier, the package information including package size information; obtaining the compartment information of the smart locker, the compartment information including compartment size information; determining the second target compartment based on the package size information and the compartment size information; and controlling the door corresponding to the second target compartment to open to complete the delivery based on the temporary operation permission.

7. The authentication method for the smart cabinet as described in claim 6, characterized in that, The step of determining the second target compartment based on the package size information and the compartment size information includes: determining a first compartment type based on the package size information and the compartment size information; determining whether there is a first available compartment in the first compartment type; if there is, selecting a first available compartment from all first available compartments of the first compartment type as the second target compartment using a preset random algorithm; if there is no first available compartment, determining a second compartment type associated with the first compartment type, and selecting a second available compartment from all second available compartments of the second compartment type as the second target compartment using the preset random algorithm.

8. A smart cabinet, characterized in that, The smart locker includes a control module; the control module is used to: acquire the user's facial biometric data and real-time movement data; input the facial biometric data and a pre-stored user identity database into a pre-trained identity verification model, so that the identity verification model outputs an identity verification result; wherein, the user identity database stores all authorized facial feature data; generate temporary operation permissions with duration constraints based on the identity verification result and the real-time movement data; and perform the required package storage and retrieval operations based on the temporary operation permissions.

9. A smart cabinet, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the authentication method for the smart cabinet as described in any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the authentication method for the smart cabinet as described in any one of claims 1-7.

Images