A hotel unmanned check-in system and method based on face recognition

Abstract

The present invention relates to the technical field of face recognition, and discloses a hotel unmanned check-in system and method based on face recognition. The hotel unmanned check-in method based on face recognition includes: obtaining multi-modal environmental data, extracting environmental risk factors, forming an environmental risk vector, and calculating a comprehensive environmental risk score; calculating a verification entropy value based on the environmental risk score, establishing a verification mode set, executing a verification entropy optimization algorithm, and forming a verification strategy table; inputting the environmental risk score into a biological cognitive synchronization control algorithm, generating a controller output value, converting it into dynamic verification parameters, executing cognitive depth adjustment, and generating a verification execution instruction set; performing a face recognition verification process according to the verification execution instruction set; through the above steps, the present invention constructs an intelligent verification system capable of perceiving the environment, evaluating risks, and adaptively adjusting, solving the problems of insufficient environmental adaptability, lack of dynamic adjustment, and missing risk prediction in the prior art.

Description

Technical Field

[0001] The present invention relates to the technical field of face recognition, and more specifically, it relates to a hotel unmanned check-in system and method based on face recognition. Background Art

[0002] Most existing face recognition verification systems adopt a static threshold determination method, which has the following technical defects:

[0003] The balance problem between security and user experience: Traditional systems use fixed security thresholds and unified verification processes, and cannot dynamically adjust verification strategies according to the actual risk level, resulting in insufficient verification and potential security risks in high-security requirements, and over-verification affecting user experience in low-risk scenarios;

[0004] Lack of environmental adaptability: Existing solutions lack the ability to quantitatively evaluate complex factors such as environmental light changes, user distance dynamics, and environmental noise, and the recognition accuracy drops significantly in complex environments such as low light and high noise;

[0005] Low resource efficiency: The verification mechanism with a fixed process still executes complete verification steps in simple scenarios, resulting in waste of computing resources;

[0006] Lack of risk prediction: It is unable to predict potential risk changes based on historical data and environmental characteristics, and lacks the ability of active defense. These defects make it difficult for existing systems to simultaneously meet the operation requirements of security, reliability, efficiency, and convenience in the hotel unattended scenario, and there is an urgent need for an intelligent verification solution that can dynamically balance security and user experience. Summary of the Invention

[0007] The present invention provides a hotel unmanned check-in system and method based on face recognition to solve the technical problems in the above related technologies.

[0008] The present invention provides a hotel unmanned check-in method based on face recognition, including:

[0009] Obtain multi-modal environmental data, extract environmental risk factors, form an environmental risk vector, and calculate a comprehensive environmental risk score;

[0010] Calculate the verification entropy value based on the environmental risk score, establish a verification mode set, execute the verification entropy optimization algorithm, and form a verification strategy table;

[0011] Input the environmental risk score into the biological cognitive synchronization control algorithm, generate a controller output value, convert it into dynamic verification parameters, execute cognitive depth adjustment, and generate a verification execution instruction set;

[0012] Execute the face recognition verification process according to the verification execution instruction set.

[0013] Furthermore, the extracted environmental risk factors include the following factors: environmental light intensity and uniformity, the rate of change of the distance between the user and the device, environmental noise level, time factor, location factor, and historical anomaly record factor.

[0014] Furthermore, the steps for calculating the comprehensive environmental risk score adopt a weighted summation formula:

[0015] ;

[0016] where is the weight value of the light factor, is the weight value of the rate of change of distance, is the weight value of the noise level, is the weight value of the time factor, is the weight value of the location factor, is the value of the historical anomaly record factor, is the weight coefficient and satisfies .

[0017] Furthermore, the calculation formula for the verification entropy value is:

[0018] ;

[0019] where represents the probability of adopting the th verification mode, is the total number of available verification modes, is the environmental risk score.

[0020] Furthermore, the set of verification modes includes: basic mode, standard mode, enhanced mode, strict mode, and highest security mode, and each mode contains different combinations of verification methods.

[0021] Furthermore, the steps for executing the verification entropy optimization algorithm obtain the optimal selection probabilities of each verification mode by solving the following optimization problem:

[0022] ;

[0023] ;

[0024] where, is the target verification entropy value, which is determined according to the environmental risk score .

[0025] Furthermore, the dynamic verification parameters include: the feature extraction depth value, the matching threshold value, the verification module activation degree vector, and the verification process complexity value.

[0026] Further, the steps of performing cognitive depth adjustment include: generating a streamlined verification instruction set in a low-risk scenario, a balanced verification instruction set in a medium-risk scenario, or a strengthened verification instruction set in a high-risk scenario according to the environmental risk level.

[0027] Further, it also includes a smooth control response step: adjusting the controller output value using the cognitive inertia parameter to generate the actual control output value, and the calculation formula is:

[0028] ;

[0029] where is the actual control output value, is the control output value at the previous moment, is the cognitive inertia coefficient.

[0030] A hotel unmanned check-in system based on face recognition, used to execute the above-mentioned hotel unmanned check-in method based on face recognition, includes:

[0031] An environmental perception and risk factor extraction module, used to obtain multi-modal environmental data, extract environmental risk factors, form an environmental risk vector, and calculate the comprehensive environmental risk score;

[0032] A verification entropy calculation and strategy generation module, used to calculate the verification entropy value based on the environmental risk score, establish a verification mode set, execute the verification entropy optimization algorithm, and form a verification strategy table;

[0033] A biological cognitive synchronization control module, used to input the environmental risk score into the biological cognitive synchronization control algorithm, generate the controller output value, convert it into dynamic verification parameters, perform cognitive depth adjustment, and generate a verification execution instruction set;

[0034] A face recognition verification module, used to execute the face recognition verification process according to the verification execution instruction set.

[0035] The beneficial effects of the present invention are as follows:

[0036] Dynamic security verification mechanism: Through the quantization model of verification entropy, the accurate matching of the security verification intensity and the environmental risk level is realized. While ensuring high security, redundant verification steps in low-risk scenarios are reduced, and the verification efficiency is increased by more than 40%;

[0037] Resource optimization efficiency: The risk-adaptive hierarchical verification architecture can dynamically select the verification mode according to the real-time risk assessment, save computing resources in conventional scenarios, and automatically enhance the verification dimension in high-risk scenarios to ensure that the security threshold is always higher than the risk level;

[0038] Intelligent Risk Prediction: The context predictive security mechanism can analyze the time series data of environmental characteristics, predict the risk trend 5 to 10 seconds in advance, and adjust the verification strategy, thereby improving the response speed of the system to sudden security incidents;

[0039] Environmental Adaptability: The biological cognitive synchronization controller can dynamically adjust the perception threshold, enabling the system to maintain a high recognition accuracy in complex environments such as low light and high noise. Brief Description of the Drawings

[0040] Figure 1 It is a flowchart of a hotel unmanned check-in method based on face recognition according to the present invention. Detailed Embodiments

[0041] Now, the subject matter described herein will be discussed with reference to exemplary embodiments. It should be understood that discussing these embodiments is only to enable those skilled in the art to better understand and thus implement the subject matter described herein. Without departing from the scope of protection of the content of this specification, changes can be made to the functions and arrangements of the elements discussed. Each example can omit, substitute, or add various processes or components as needed. Additionally, the features described in some examples can also be combined in other examples.

[0042] Embodiment 1, A hotel unmanned check-in method based on face recognition, as Figure 1 shown, includes the following steps:

[0043] Step 101: Convert the environmental data collected by the sensor into an environmental risk score;

[0044] Obtain multi-modal environmental data: A multi-modal sensor array (visible light camera, infrared camera, ambient light sensor, distance sensor, noise sensor, etc.) collects real-time data streams of the verification environment to form an original environmental data set;

[0045] Extract environmental risk factors: The environmental feature extraction unit receives the original environmental data set and extracts the following risk factor values:

[0046] Environmental light intensity and uniformity , representing the lighting state at position at time ;

[0047] Rate of change of the distance between the user and the device , representing the change speed of the distance between the user and the verification device per unit time;

[0048] Environmental noise level , representing the noise decibel value of the verification environment Time factor , representing the position factor of the risk weight value at the current time period , representing the historical anomaly record factor of the risk weight value in the current area , representing the statistical value of historical anomaly events at this location.

[0049] Generate the environmental risk vector: The normalization unit receives the risk factor values and outputs the normalized environmental risk vector:

[0050]

[0051] Among them , , , , , are the normalized weight values corresponding to environmental light intensity and uniformity, the rate of change of the distance between the user and the device, environmental noise level, time factor, location factor, and historical anomaly record factor respectively;

[0052] Aggregate the risk score: The risk assessment function receives the environmental risk vector , and outputs the comprehensive environmental risk score as a single numerical value , and the calculation formula is:

[0053]

[0054] In the formula, is the weight coefficient of each risk factor, which is automatically optimized and adjusted through machine learning, and .

[0055] Comprehensive environmental risk score is a scalar value in the range of [0, 1], representing the overall risk level of the current environment, and serves as the input for subsequent verification entropy calculation and control parameter generation.

[0056] Different from the traditional method, this step realizes the precise quantitative evaluation of the verification environment through the dynamic weighted combination of risk factors, providing a mathematical basis for the adaptive adjustment of verification strategies.

[0057] Step 102: Receive the environmental risk score as input and output the optimal verification strategy for the current environment;

[0058] Calculate the verification entropy value: The verification entropy calculation module receives the environmental risk score of Step 101 , and calculates the verification entropy value in the current situation , representing the complexity and strictness required for the system verification process:

[0059]

[0060] In the formula, represents the adoption of the The probability of a verification mode is the total number of available verification modes is the environmental risk score. The verification entropy The higher the value, the more complex and strict the verification process is required.

[0061] Integrate the verification mode library: The verification mode selection unit constructs a set of verification modes, including multiple verification modes and their corresponding security strengths and user experience consumption values:

[0062] Basic mode: Only use facial feature point comparison;

[0063] Standard mode: Facial feature point comparison + liveness detection;

[0064] Enhanced mode: Facial feature point comparison + liveness detection + multi-angle verification;

[0065] Strict mode: Facial feature point comparison + liveness detection + multi-angle verification + behavior feature analysis;

[0066] Highest security mode: Full-dimensional biometric fusion verification;

[0067] Optimize the verification strategy combination: The strategy optimization unit receives the verification entropy value and the set of verification modes, and executes the verification entropy optimization algorithm to determine the optimal verification strategy combination plan. This algorithm obtains the optimal selection probability of each verification mode by solving the following optimization problem:

[0068]

[0069]

[0070] where represents finding the variable value combination that minimizes the objective function, that is, finding the probability distribution parameters that make the actual verification entropy closest to the target verification entropy , ,…, , represents the th selection probability of the verification mode is the target verification entropy value (determined according to the risk score ). This algorithm finds the probability distribution that makes the actual verification entropy closest to the target verification entropy through iterative calculation.

[0071] Form a verification strategy table: The verification strategy generation unit receives the optimized verification mode probability distribution and generates a complete verification strategy table , including: the selection probability values of each verification mode The execution order table of each verification module and the threshold parameter table of each verification module; The verification strategy table It is a structured data set that contains all the configuration information required for the system to execute the verification process.

[0072] Different from traditional fixed verification strategies, this step realizes the dynamic matching of the verification strategy and environmental risks by verifying entropy, ensuring that the system adopts the most suitable verification configuration under different risk levels.

[0073] Step 103: Convert the environmental risk score data into dynamic verification parameters to achieve real-time regulation of the verification process;

[0074] Form the controller output: The biological cognitive synchronization control algorithm receives the real-time environmental risk score data stream , and generates the controller output value , which determines the cognitive depth of the system verification process. The core mathematical model of the controller is:

[0075]

[0076] In the formula, is the output value of the controller at time ; is the environmental risk score at time ; is the risk change rate, indicating the degree of risk sharpness; is the cumulative risk value, indicating the long-term risk state; , , are the first, second, and third control parameters, respectively representing the response intensities of the system to the risk change rate, the current risk value, and the cumulative risk.

[0077] Convert the verification parameters: The parameter mapping unit receives the controller output value , and generates the following dynamic verification parameters:

[0078] Feature extraction depth value , which controls the fineness of face feature extraction;

[0079] Matching threshold value , which controls the strictness of feature matching;

[0080] Verification module activation degree vector , which determines the participation degree of different verification modules;

[0081] Verification process complexity value , which controls the number of steps in the verification process;

[0082] Among them , , , They are the first, second, third, and fourth parameter mapping functions respectively, which convert the controller output value into the corresponding verification parameter value. These parameters constitute the dynamic configuration data set of the verification process.

[0083] Execute cognitive depth adjustment: The cognitive depth adjustment unit receives the verification parameters, adjusts the cognitive depth of the system according to the environmental risk level, and generates specific verification execution instructions:

[0084] The streamlined verification instruction set in a low-risk scenario: Reduce the cognitive depth and the number of verification steps;

[0085] The balanced verification instruction set in a medium-risk scenario: Maintain a moderate cognitive depth;

[0086] The enhanced verification instruction set in a high-risk scenario: Increase the cognitive depth and the number of verification steps;

[0087] Smooth control response: The cognitive inertia processing unit receives the original controller output value , applies the cognitive inertia parameter, and generates the actual control output value considering cognitive inertia:

[0088]

[0089] In the formula, is the actual control output after adding cognitive inertia, is the control output at the previous moment, is the cognitive inertia coefficient, which is used to balance the response speed and stability of the system to risk changes.

[0090] The finally output verification execution instruction set contains the complete verification process configuration information, which will be used by the verification execution unit to guide the actual face recognition verification process.

[0091] Different from the traditional fixed control method, this step introduces a differential-integral control mechanism similar to biofeedback regulation, enabling the system to smoothly adjust the verification parameters, avoiding the sudden change behavior of the traditional system when the risk changes, and improving the user experience.

[0092] Step 104: Convert the historical verification data into a risk prediction model, output the future risk prediction value, and achieve anticipatory safety adjustment.

[0093] Construct spatio-temporal data records: The data acquisition unit collects various historical data of the verification environment and constructs a spatio-temporal feature database, including: Time feature data: period identifier, date type, season mark, holiday flag, etc. Location feature data: area code, entrance / exit type identifier, surrounding environment description, etc. Historical risk record data: verification failure rate statistics, attack detection rate value, abnormal behavior frequency table, etc. System response performance data: average verification time record, resource occupancy rate statistics, user waiting duration table, etc.;

[0094] Extract the spatio-temporal risk law: The spatio-temporal pattern mining unit receives the spatio-temporal feature database, applies the Bayesian conditional probability model, and outputs the risk distribution law matrix:

[0095]

[0096] Wherein, represents the risk probability distribution under specific time type and location type conditions; represents the probability of a certain time and location combination occurring under a specific risk level; represents the prior probability of the risk level; represents the marginal probability of the time and location combination. The risk distribution law matrix contains the statistical laws of the risk level under different spatio-temporal conditions.

[0097] Generate risk trend prediction: The risk prediction unit receives the current and historical risk data sequences and spatio-temporal information, executes the time series prediction algorithm, and outputs the future risk prediction value:

[0098]

[0099] Wherein, is the predicted risk value after time; is the time series prediction function; , ,…, are the risk values at the current and past consecutive time points; , are the time and location features respectively. The future risk prediction value is a scalar in the range of [0, 1], indicating the expected future risk level of the system.

[0100] Execute anticipatory adjustment: The risk pre-control unit receives the risk prediction value at the next future time point and the current risk score , calculates the expected risk change , and outputs different system adjustment instructions according to the change expectation:

[0101] Risk increase pre-adjustment instruction: When ( is the preset threshold), initiate the safety resource allocation in advance to prepare for the risk increase;

[0102] Risk decrease pre-adjustment instruction: When , release the excess safety resources in advance to optimize the user experience;

[0103] These pre-adjustment instructions will be received by the system resource management unit to pre-adjust the system configuration before the actual risk changes, enabling an early response.

[0104] Different from traditional passive response methods, this step constructs a risk trend prediction model by analyzing the spatio-temporal patterns of historical data, enabling the system to anticipate potential risk changes and make adjustments in advance, thereby improving the system's security response speed and user experience.

[0105] Technical effects of this embodiment:

[0106] The adaptive security verification system constructed based on the "Bio-Cognitive Synchronization Theory" in this embodiment achieves the following technical effects:

[0107] Synchronous improvement of security and user experience: Through the verification entropy optimization model and the bio-cognitive synchronization controller, the system can dynamically adjust the verification strategy according to the actual risk. While maintaining equivalent security, the verification process is smoother, the average user waiting time is reduced by 78%, and the user satisfaction is significantly improved.

[0108] Efficient utilization of system resources: Since the verification intensity is precisely matched with the actual risk, system resources are no longer wasted by excessive verification. The resource occupancy is reduced by 52%, while the processing capacity is improved and the peak user concurrency is increased by 3 times.

[0109] Adaptability to complex environments: The system can perceive and adapt to different environmental conditions (such as light, noise, etc.), and maintain a high recognition accuracy even under harsh conditions. The environmental adaptability is 65% higher than that of traditional fixed-threshold systems.

[0110] Predictive security protection: Through the context predictive security mechanism, the system can predict risk changes and make adjustments in advance, proactively coping with potential threats, advancing the security response time by about 2 - 5 minutes, and significantly reducing security hazards caused by lagged responses.

[0111] Continuous evolution of security awareness: The system continuously optimizes the risk assessment model and control parameters by learning historical verification data, and the recognition accuracy steadily improves with the usage time. The misrecognition rate after one year of use is 45% lower than the initial state.

[0112] This embodiment fundamentally changes the verification mode of the traditional face recognition system of "static threshold, fixed process", realizing the technical paradigm shift from "mechanical comparison" to "bio-cognition", and providing an optimal balance solution for security and user experience in the hotel unmanned check-in scenario.

[0113] An application example of Embodiment 1 is as follows:

[0114] Deployment tests were conducted in the unmanned check-in system of a five-star hotel chain. The hotel has 150 guest rooms with an average daily occupancy rate of 85%. The flow of people is large and the customer composition is complex (including business travelers, leisure tourists, conference participants, etc.). The system is deployed at the entrance of the hotel lobby and in the elevator hall on each floor, and a total of 6 verification terminal devices are included.

[0115] Characteristics of the system deployment environment:

[0116] Large variation in lighting conditions: The lighting in the lobby ranges from 300 lux in the morning to 800 lux at night, and there is a mixed environment of dynamic natural light and artificial light;

[0117] Unstable noise level: Ranging from 35 dB during quiet periods to 75 dB during peak periods;

[0118] Variation in user distance: The distance between the user and the device ranges from 0.3 meters to 2 meters;

[0119] Uneven distribution of security risks: There are obvious differences in the security risk levels in different time periods and regions;

[0120] To visually display the differences in security risks in different regions and time periods, the system established a basic risk rating table as a reference benchmark for dynamic risk assessment. The basic risk ratings of different regions and time periods in the hotel are shown in Table 1:

[0121] Table 1: Basic risk ratings of different regions and time periods in the hotel

[0122]

[0123] The risk rating values in Table 1 range from 0 to 1, and the larger the value, the higher the risk.

[0124] Implementation process example:

[0125] Example of environmental perception and risk factor extraction:

[0126] Taking a business traveler using the unmanned check-in system in the hotel lobby at 19:32 in the evening as an example, the actual execution process of each step is shown.

[0127] The original environmental data collected by the multi-modal sensors, the original environmental data of a check-in verification event is shown in Table 2:

[0128] Table 2: Example of the original environmental data of a check-in verification event

[0129]

[0130] The results of environmental risk factor extraction are shown in Table 3:

[0131] Table 3: Numerical values of the extracted environmental risk factors

[0132]

[0133] Calculation of environmental risk vector and comprehensive score:

[0134] Environmental risk vector:

[0135]

[0136] Comprehensive environmental risk weight coefficient based on dynamic learning:

[0137]

[0138] Calculated comprehensive environmental risk score:

[0139]

[0140] This comprehensive environmental risk score indicates that the current verification environment is at a medium risk level, between low risk and high risk, and appropriate security verification measures need to be taken.

[0141] Example of verification entropy calculation and verification strategy generation:

[0142] Calculation of verification entropy value:

[0143] Based on the aforementioned environmental risk score , the system calculates the verification entropy value in the current situation. The initial probability distribution and entropy contribution of the verification mode are shown in Table 4:

[0144] Table 4: Initial probability distribution and entropy contribution of verification mode

[0145]

[0146] Calculated initial verification entropy value:

[0147]

[0148] Results of verification strategy optimization:

[0149] According to the current medium risk level, the target verification entropy . The system executes the verification entropy optimization algorithm to solve the optimization problem:

[0150]

[0151]

[0152] The optimized probability distribution of the verification mode is shown in Table 5:

[0153] Table 5: Optimized probability distribution of verification mode

[0154]

[0155] Optimized verification entropy value: , close to the target verification entropy value of 0.75.

[0156] Verification strategy table generation:

[0157] The system generates a complete verification strategy table according to the optimized probability distribution as shown in Table 6:

[0158] Table 6: Final generated verification strategy table (partial key parameters)

[0159]

[0160] This verification strategy table provides the basic configuration information for the subsequent biocognitive synchronization controller and can be further adjusted according to real-time situations.

[0161] Implementation example of biocognitive synchronization controller:

[0162] The system inputs the environmental risk score data into the biocognitive synchronization controller for dynamic adjustment of the verification process.

[0163] Controller output value calculation:

[0164] In this case, the system not only considers the current risk score , but also considers historical data:

[0165] Risk change rate: (Risk is slightly increasing);

[0166] Cumulative risk value: (Risk accumulation value in the past 30 minutes);

[0167] Control parameters used: ;

[0168] Calculate the controller output value:

[0169]

[0170] The controller parameters and output are shown in Table 7:

[0171] Table 7: Controller parameter and output calculation

[0172]

[0173] Dynamic verification parameter conversion:

[0174] The controller output value is converted into specific dynamic verification parameters as shown in Table 8:

[0175] Table 8: Controller Output Converted to Verification Parameters

[0176]

[0177] Execute Cognitive Depth Adjustment:

[0178] The system performs cognitive depth adjustment based on the calculated verification parameters and generates verification execution instructions as shown in Table 9:

[0179] Table 9: Execution Instructions after Cognitive Depth Adjustment

[0180]

[0181] Smooth Control Response:

[0182] To avoid sudden changes in verification parameters, the system applies cognitive inertia parameters for smoothing:

[0183] Cognitive Inertia Coefficient ;

[0184] Control Output at the Previous Moment ;

[0185] Actual Control Output Value:

[0186]

[0187] The control output after smoothing is slightly reduced, making the system response more stable, avoiding sudden changes in verification strategies, and enhancing the user experience.

[0188] Face Recognition Verification Process Execution Example:

[0189] Based on the verification execution instruction set generated in the foregoing steps, the system performs face recognition verification on the user, and the execution record of the verification process is shown in Table 10:

[0190] Table 10: Execution Record of Face Recognition Verification Process

[0191]

[0192] Compared with the traditional fixed verification process, the dynamic adjustment strategy adopted in this example effectively enhances the user experience while maintaining security:

[0193] Although more verification steps are executed, the system only adds necessary verification links according to real-time risk assessment;

[0194] The live detection process gives clear guidance to the user, avoiding user operation errors;

[0195] The verification intensity is moderate, neither too loose nor too strict, and matches the current environmental risk level.

[0196] Verification of Technical Effects:

[0197] In the actual deployment test of this embodiment in a five-star hotel chain in Shanghai, two of the most critical technical effects were selected for verification: the balanced optimization between security and user experience, and the efficient utilization of system resources.

[0198] Verification of the Balanced Effect between Security and User Experience:

[0199] Test Method: Compare the performance of the traditional fixed-strategy face recognition system and this embodiment in the same scenario, and collect data for 30 days, including indicators such as false recognition rate, user waiting time, and user feedback. The comparison data of security and user experience are shown in Table 11:

[0200] Table 11: Comparison Data of Security and User Experience

[0201]

[0202] Data Analysis:

[0203] While improving security, this embodiment significantly improves the user experience;

[0204] For scenarios with different risk levels, the system can dynamically adjust the verification strategy, significantly shortening the average verification time;

[0205] The significant reduction in the user operation retry rate indicates a significant improvement in the system's environmental adaptability and clearer guidance;

[0206] The user's subjective satisfaction has been greatly improved, and the number of complaints has decreased by 76%.

[0207] Verification of System Resource Utilization Efficiency:

[0208] Test Method: Test the resource occupancy and processing capabilities of the two systems during peak and off-peak periods respectively, and collect data such as CPU usage rate, memory occupancy, and concurrent processing capabilities. The comparison data of system resource utilization efficiency are shown in Table 12:

[0209] Table 12: Comparison Data of System Resource Utilization Efficiency

[0210]

[0211] Data Analysis:

[0212] This embodiment significantly reduces the overall system resource occupancy through the strategy of "allocating resources on demand";

[0213] In a low-risk environment, the resource occupancy reduction is the most obvious (57%), fully demonstrating the system's intelligent scheduling ability;

[0214] The peak concurrent processing capacity is increased by 175%, which means that under the same hardware conditions, the system service capacity is greatly improved;

[0215] The system response latency is reduced by 67.1%, further improving the user interaction experience.

[0216] Comprehensive analysis shows that this implementation method successfully solves the contradiction between security and user experience in traditional face recognition systems through the adaptive security verification system of biological cognitive synchronization, and significantly improves the system resource utilization efficiency.

[0217] Implementation method 2:

[0218] This implementation method is applied to the hotel unmanned check-in scenario based on face recognition. On the basis of implementation method 1, it solves the collaborative decision-making problem in a large-scale distributed verification environment. The technical requirements include:

[0219] Support distributed collaborative verification of multiple entrances and multiple terminals to ensure security and user experience;

[0220] Solve the asynchronous decision-making coordination problem of different levels of verification nodes (such as the main entrance of the lobby, the entrance of the floor, the room access control, etc.);

[0221] Reduce the computing bottleneck and communication latency caused by the centralized decision-making architecture;

[0222] Realize the intelligent allocation and load balancing of verification resources;

[0223] Ensure the system availability and consistency in case of network fluctuations and partial node failures.

[0224] Traditional verification systems generally adopt a synchronous decision-making architecture, where all verification decisions are centrally processed on the same time scale and decision level. It is difficult to support the high-concurrency verification requirements in a distributed environment and is prone to single-point failures. Although implementation method 1 introduces an adaptive verification mechanism of biological cognitive synchronization, it still does not fully solve the collaborative decision-making problem between different verification nodes in a distributed environment.

[0225] A hotel unmanned check-in method based on face recognition in implementation method 2 includes the following steps:

[0226] Step 201: Environment perception and risk factor extraction, generating a distributed comprehensive environment risk scoring matrix;

[0227] Obtain environmental sensor data from multiple verification nodes, including light intensity data, background noise data, crowd density data, verification terminal status data, and network connection quality data. According to the obtained multi-modal environmental data, extract environmental risk factors respectively, and generate a distributed comprehensive environment risk scoring matrix:

[0228]

[0229] Among them: represents the distributed comprehensive environment risk scoring matrix, which contains the environment risk scores of all verification nodes; represents the th comprehensive environment risk score of the verification node, with a value range of [0, 1]. The larger the value, the higher the environment risk; represents the total number of verification nodes in the system.

[0230] Step 202: Generate a multi-time scale asynchronous decision model;

[0231] Construct the topology structure data of the hotel verification system, including verification node distribution data, inter-node connection relationship data, and node computing power data; obtain the distributed comprehensive environment risk scoring matrix in Step 201 ;

[0232] After analyzing and processing the topology structure data of the hotel verification system, generate a multi-time scale asynchronous decision model applicable to the distributed verification environment , and this model includes the following components:

[0233]

[0234] Among them, the time scale mapping table is calculated and generated based on the environment risk scoring matrix:

[0235]

[0236] Specific time scale parameters are adjusted through the environment risk matrix:

[0237]

[0238]

[0239]

[0240] Among them from Step 201 , and are weight coefficients, , and are basic time scale parameters.

[0241] This multi-time scale asynchronous decision model decomposes the verification decision into three different time scale levels and provides a coordination method between levels, enabling the system to process according to the nature of the decision task at the appropriate level and time scale.

[0242] Step 203: Formation of an edge-cloud collaborative hierarchical verification architecture;

[0243] Obtain the verification node hardware specification data, network connection data, verification task historical execution data, and the multi-time scale asynchronous decision-making model generated in step 202 and, after analysis and processing, form an edge-cloud collaborative hierarchical verification architecture description file :

[0244]

[0245] where represents the edge-cloud collaborative hierarchical verification architecture description file, which stipulates the hierarchical processing method of verification tasks;

[0246] represents the node capability matrix, where represents the th node's capability vector, containing computing, storage, and network capability parameters, represents the total number of nodes in the system; represents the task requirement matrix, where represents the th type of task's resource requirement vector, represents the total number of task types in the system; represents the three-layer physical architecture deployment plan, specifying the deployment locations of various verification components in the edge layer, fog computing layer, and cloud computing layer; represents the resource allocation matrix, specifying the optimal matching relationship between tasks and nodes; represents the communication network configuration, including the configuration parameters of three types of communication channels; represents the fault recovery process, including node status monitoring, task migration, and data backup strategies.

[0247] Resource allocation matrix According to the task allocation function from calculated as:

[0248]

[0249] where is the th decision task, and are respectively the level and processing unit of the th node, is the allocation threshold.

[0250] This edge-cloud collaborative hierarchical verification architecture effectively supports verification tasks in a distributed environment by allocating verification tasks of different complexities to corresponding computing levels, optimizing the utilization efficiency of system resources.

[0251] Step 204: Hierarchical verification entropy calculation and collaborative strategy generation;

[0252] Receive the distributed integrated environment risk scoring matrix generated in step 201 , the multi-time scale asynchronous decision-making model generated in step 202 , and the historical verification result data set, calculate the verification entropy values of each level and generate a hierarchical collaborative verification strategy table applicable to the distributed environment :

[0253]

[0254] Among them, the verification mode selection probability matrix of each level is associated with the environmental risk and time scale:

[0255]

[0256]

[0257]

[0258] Where is the average risk score, from ; , and from in ; represents the energy consumption of the verification mode in level ; , and are adjustment parameters.

[0259] The hierarchical verification entropy calculation is based on the principle of information theory, quantifying the uncertainty of verification tasks at each level and providing a theoretical basis for verification strategy selection. The formula for calculating the verification entropy value is:

[0260]

[0261] Where: represents the verification entropy value of level , quantifying the uncertainty of the verification decision at this level; represents the selection probability of the verification mode in level , and the sum of all probabilities is 1; Represents the hierarchy The number of optional verification modes in; Represents the natural logarithm; Represents the hierarchy All in The verification modes are summed up.

[0262] By adjusting the verification mode selection probability to make the verification entropy value close to a specific target value, the effect of balancing verification complexity and environmental risk can be achieved.

[0263] Step 205: Implementation of the distributed biocognitive synchronization controller;

[0264] Receive the distributed comprehensive environmental risk scoring matrix generated in step 201 , the multi-time scale asynchronous decision-making model generated in step 202 , the edge-cloud collaborative hierarchical verification architecture description file generated in step 203 , and the system operation feedback data to implement a biocognitive synchronization controller network that can adapt to the distributed environment :

[0265]

[0266] Among them: Represents the distributed biocognitive synchronization controller network, responsible for the control and coordination of the entire distributed verification system; Represents the set of edge controllers, where Represents the Parameter configuration of the th edge controller; Represents the set of area controllers, where Represents the Parameter configuration of the th area controller; Represents the parameter configuration of the global controller, responsible for global policy formulation; Represents the inter-level control coordination matrix, used to coordinate the outputs of different-level controllers; Represents the controller parameter adaptive adjustment strategy, including learning rate, iteration step size, and convergence condition; Represents the control instruction distribution and synchronization rule table, which stipulates the generation, distribution, and synchronization methods of control instructions; Represents the control conflict resolution strategy table, which defines conflict detection rules and resolution methods.

[0267] Inter-level control coordination matrix Based on the time scale And the resource allocation matrix Build:

[0268]

[0269] wherein represents the time scale obtained from the obtained hierarchy , from the resource allocation matrix is the normalization coefficient. The controller parameter adjustment strategy is based on the environmental risk score:

[0270]

[0271] wherein and from , represents the set of nodes included in the area , is the controller sensitivity parameter.

[0272] The controllers at each level adopt mathematical models with different complexities to adapt to the decision-making requirements under the corresponding time scales.

[0273] Step 206: Implement the distributed verification consensus mechanism;

[0274] Receive the edge-cloud collaborative hierarchical verification architecture description file generated in Step 203, the system topology data, the historical verification result data set, and the consensus performance requirement parameters, and implement a consensus mechanism to solve the consistency problem of multiple verification nodes in a distributed environment, forming a distributed verification consensus system :

[0275]

[0276] wherein: represents the distributed verification consensus system, ensuring the consistency of verification results in a distributed environment; represents the verification status vector, describing the key status and its data structure of the verification process; represents the lightweight consensus protocol, including the state propagation rule, the voting algorithm, and the conflict resolution steps; represents the consensus parameter set, including the optimization parameters and thresholds of the consensus protocol; represents the progressive consensus mechanism description, including the execution conditions and processes of the preliminary, enhanced, and final consensus stages; represents the distributed verification state machine, describing the state transition logic in the verification process; represents the verification result consistency proof generation function, used to generate a verifiable consistency proof.

[0277] The consensus parameter set is optimized according to the deployment plan and network configuration:

[0278]

[0279] The phase transition conditions of the progressive consensus mechanism are related to resource allocation:

[0280]

[0281] Among them and from the deployment plan and network configuration in represents the set of nodes that have participated in the consensus, represents the node at the level resource allocation value, is the consensus threshold for phase .

[0282] Step 207: Group verification collaborative learning is realized;

[0283] Receive the hierarchical collaborative verification policy table generated in step 204 , the verification experience data of each verification node, the user behavior data set, and the environmental condition verification performance correlation data, and implement the knowledge sharing and collaborative evolution mechanism among the verification nodes to form a group verification collaborative learning system :

[0284]

[0285] Verification performance evaluation function Calculate according to the policy parameters:

[0286]

[0287] State-action value function Integrate the policy parameters and verification entropy:

[0288]

[0289] Among them from the verification mode selection probability matrix of the corresponding level in represents performing the action in the state and selecting the verification mode of level reward.

[0290] Anomaly detection and adaptive adjustment rules are based on the verification policy consistency maintenance policy:

[0291] ​

[0292] Among them from , is the verification result consistency maintenance strategy parameter at the hierarchical level, indicating the average verification mode probability distribution at the hierarchical level in the group, and is the adjustment threshold.

[0293] Group verification collaborative learning adopts federated learning and differential privacy technologies, allowing each verification node to share knowledge while protecting data privacy.

[0294] Step 208: Execute the face recognition verification process.

[0295] Receive the hierarchical collaborative verification policy table generated in step 204 , the distributed biological cognitive synchronization controller network generated in step 205 , the distributed verification consensus system generated in step 206 , the group verification collaborative learning system generated in step 207 , as well as the face image data, user interaction data, and environmental sensor data collected in real time, and execute the complete face recognition verification process to generate a verification result data packet :

[0296]

[0297] Among them, the verification confidence score synthesizes the outputs of each system:

[0298]

[0299] Among them from , from , from is the consistency measure, from is the performance evaluation function, to is the weight coefficient, is the currently activated verification policy, is the hierarchical level in the verification mode

[0300] The verification consensus proof is generated through the consensus system:

[0301]

[0302] Among them , and are all from .

[0303] The verification feedback data integrates the status information of each subsystem:

[0304]

[0305] Among them, represents the set of hierarchical collaborative verification strategy control parameters; represents the real-time configuration parameters of the distributed biological cognitive synchronization controller; represents the distributed verification consensus progress matrix; represents the group verification collaborative Q function.

[0306] In this step, according to the hierarchical collaborative verification strategy and distributed control instructions generated in the previous step, the complete face recognition verification process is executed in a distributed environment, which can utilize the collaborative ability and consensus mechanism of multiple verification nodes to provide more reliable and efficient verification results.

[0307] The distributed verification system implemented based on the "asynchronous hierarchical biological cognitive collaborative model" realizes the following specific technical effects through the organic combination of the above steps 201 to 208:

[0308] High concurrency processing ability: The multi-time scale asynchronous decision-making model in step 202 and the edge-cloud collaborative hierarchical verification architecture in step 203 work together, enabling the system to allocate verification tasks to appropriate computing levels and time scales according to the urgency and complexity of the verification tasks. In actual tests, compared with the traditional synchronous decision-making architecture, the peak processing ability of this distributed verification system has increased by 720%, and the average response time has been shortened from the original 2.5 seconds to 0.375 seconds (a reduction of 85%), which is particularly suitable for large-scale concurrent verification scenarios during the peak hotel check-in period.

[0309] Collaboration between global and local security policies: The hierarchical verification entropy calculation and collaborative strategy generation in step 204, combined with the distributed biological cognitive synchronization controller in step 205, realize the dynamic balance between global security policies and local environmental conditions. During the verification process, the system can maintain equivalent security (without reducing the verification accuracy rate) while dynamically adjusting the verification strategy according to local environmental risks, making the verification process smoother, and the average waiting time of users has been reduced from the original 8 seconds to 4.8 seconds (a reduction of 40%).

[0310] Distributed Consistency and Reliability: The distributed verification consensus mechanism in Step 206 solves the data consistency problem in a distributed environment, ensuring that the system can still operate reliably as a whole even when some nodes fail. After a three-month actual deployment test, the system availability has been improved from the traditional 99.9% (about 43 minutes of downtime per month) to 99.999% (about 26 seconds of downtime per month), significantly reducing service interruptions caused by system failures.

[0311] Self-adaptability and Evolutionary Ability: The group verification collaborative learning system in Step 207 enables each verification node to continuously optimize the verification strategy based on the shared knowledge base and their respective verification experiences. Long-term operation tests show that the verification accuracy of the system continues to improve with the increase in usage time. The misidentification rate after two years of use is reduced by 72% compared to the initial state (from the initial 2.5% to 0.7%), demonstrating the self-evolutionary ability of the system.

[0312] Resource Utilization Efficiency: The combination of the multi-time-scale asynchronous decision-making model in Step 202 and the edge-cloud collaborative architecture in Step 203 realizes the dynamic allocation of verification tasks among different computing nodes according to complexity and urgency. In the actual deployment test, the overall resource utilization rate has increased from the original 52% to 86% (a 65% increase), the system power consumption has decreased from the original average of 120 watts to 50 watts (a 58% decrease), and at the same time, the scalability of the system has been significantly enhanced, being able to easily adapt to various scale requirements from small hotels to large hotel chains.

[0313] The above technical effects show that through the organic combination of Steps 201 to 208, this embodiment realizes the technical leap from single-node intelligence to group collaborative intelligence, providing an efficient, reliable, and scalable face recognition verification solution for the hotel self-check-in scenario. The specific technical components generated by each step (such as verification architecture description files, asynchronous decision-making models, hierarchical verification strategy tables, etc.) cooperate with each other to form a complete technical system, jointly supporting the realization of the above technical effects.

[0314] The embodiments of the present invention have been described above, but these embodiments are not limited to the above specific implementation manners. The above specific implementation manners are only illustrative and not restrictive. Under the inspiration of this embodiment, those of ordinary skill in the art can also make more equivalent embodiments in various forms, all of which fall within the protection scope of this embodiment.

Claims

1. A hotel unattended check-in method based on face recognition, characterized in that: The following steps are involved: Acquire multimodal environmental data, extract environmental risk factors, form environmental risk vectors, and calculate comprehensive environmental risk scores; Calculate the verification entropy value based on the environmental risk score, establish a verification mode set, execute the verification entropy optimization algorithm, and form a verification strategy table; The environmental risk score is input into the biological cognitive synchronous control algorithm, the controller output value is generated, converted into a dynamic verification parameter, the cognitive depth adjustment is performed, and a verification execution instruction set is generated; wherein the biological cognitive synchronous control algorithm receives the real-time environmental risk score data stream , generating the controller output value , which determines the cognitive depth of the system verification process. The core mathematical model of the controller is: ； In the formula, For the controller at time The output value of For time Environmental risk scores; is the risk change rate, indicating the severity of the risk; is the cumulative risk value, indicating the long-term risk status; , , are the first, second and third control parameters, which respectively represent the response intensity of the system to the risk change rate, current risk value and cumulative risk; Executing the face recognition verification process according to the verification execution instruction set specifically includes: obtaining the feature extraction depth value, matching threshold value, verification module activation vector and verification process complexity value contained in the verification execution instruction set; controlling the precision of face feature extraction according to the feature extraction depth value, controlling the strictness of feature matching according to the matching threshold value, and determining the participation degree of different verification modules according to the verification module activation vector; dynamically selecting a verification mode suitable for the current environmental risk score from the verification mode set based on the verification process complexity value; and performing face recognition verification on the user according to different verification modes; The hotel room access control system is controlled based on the identification and verification results to allow or deny user check-in.

2. A hotel unattended check-in method based on face recognition according to claim 1, characterized in that: Extracting environmental risk factors includes extracting the following factors: ambient light intensity and uniformity, rate of change of distance between the user and the device, ambient noise level, time factor, location factor, and historical abnormal record factor.

3. A hotel unattended check-in method based on face recognition according to claim 2, characterized in that: The steps to calculate the comprehensive environmental risk score use a weighted summation formula: ； in is the illumination factor weight value, is the distance change rate weight value, is the noise level weight value, is the time factor weight value, is the position factor weight value, is the historical abnormal record factor value, is the weight coefficient and satisfies .

4. The hotel unattended check-in method based on face recognition according to claim 3, characterized in that: The calculation formula for verification entropy value is: ； in Indicates the use of The probability of a verification mode, is the total number of available authentication modes, Score environmental risks.

5. The hotel unattended check-in method based on face recognition according to claim 4, characterized in that: The verification mode set includes: basic mode, standard mode, enhanced mode, strict mode and highest security mode, and each mode contains a different combination of verification methods.

6. A hotel unattended check-in method based on face recognition according to claim 5, characterized in that: The steps of executing the verification entropy optimization algorithm are to obtain the optimal selection probability of each verification mode by solving the following optimization problem: ； ； in, Verify entropy value for target, based on environmental risk score Sure.

7. A hotel unattended check-in method based on face recognition according to claim 6, characterized in that: Dynamic verification parameters include: feature extraction depth value, matching threshold value, verification module activation vector and verification process complexity value.

8. The hotel unattended check-in method based on face recognition according to claim 7, characterized in that: The steps of performing cognitive depth adjustment include: generating a streamlined verification instruction set for low-risk scenarios, a balanced verification instruction set for medium-risk scenarios, or an enhanced verification instruction set for high-risk scenarios based on the environmental risk level.

9. The hotel unattended check-in method based on face recognition according to claim 8, characterized in that: It also includes a smooth control response step: using cognitive inertia parameters to adjust the controller output value to generate an actual control output value, calculated as: ； in is the actual control output value, is the control output value at the previous moment, is the cognitive inertia coefficient, is the original controller output value, i.e., the controller at time The output value of .

10. A hotel unmanned check-in system based on face recognition, characterized in that: The method for detecting an unattended hotel stay based on face recognition according to any one of claims 1 to 9 comprises: Environmental perception and risk factor extraction module, which is used to obtain multimodal environmental data, extract environmental risk factors, form environmental risk vectors, and calculate comprehensive environmental risk scores; Verification entropy calculation and strategy generation module, used to calculate the verification entropy value based on the environmental risk score, establish a verification mode set, execute the verification entropy optimization algorithm, and form a verification strategy table; A bio-cognitive synchronous control module is used to input the environmental risk score into the bio-cognitive synchronous control algorithm, generate the controller output value, convert it into a dynamic verification parameter, perform cognitive depth adjustment, and generate a verification execution instruction set; The face recognition verification module is used to execute the face recognition verification process according to the verification execution instruction set.

Images