Augmented reality glasses system, glasses and storage medium

Abstract

The invention relates to the technical field of AR glasses, in particular to an augmented reality glasses system, glasses and a storage medium. Iris texture and living body features are obtained through an iris feature extraction module, and meanwhile a heartbeat feature analysis module collects heart rate signals and extracts variability features; the living body anti-counterfeiting module performs multi-dimensional fusion analysis on the two biological characteristics, and evaluates the authenticity of the characteristics to verify the identity of the user; the dynamic key generation module generates an initial key by using the verified feature data, and updates the initial key according to a preset frequency; and the data encryption module encrypts the user data by using the dynamic key. According to the method, the anti-counterfeiting capability is improved through biological feature co-verification, the encryption strength is enhanced by using the physiological feature to drive the key to dynamically update, and the security and reliability of the AR glasses system are improved.

Description

Technical Field

[0001] The present application relates to the technical field of AR glasses, and in particular to an augmented reality glasses system, glasses, and storage medium. Background Art

[0002] With the rapid development of augmented reality technology, AR glasses have become an important platform for human-computer interaction. During the application of AR glasses, users' personal data security and privacy protection issues are becoming increasingly prominent, especially in the identity authentication and data transmission stages, which require the establishment of reliable security protection mechanisms.

[0003] Current AR glasses generally use traditional biometric authentication technologies, such as iris recognition or facial recognition. Data encryption relies primarily on static keys and fixed encryption algorithms, with key information stored in hardware security modules.

[0004] However, the single biometric feature of the existing technology is easily forged and attacked; the static key is easy to steal and has a long update cycle; this seriously affects the security and reliability of the AR glasses system, and this situation needs further improvement. Summary of the Invention

[0005] To address the problems of existing biometric features being easily forged and attacked, static keys being easily stolen and having a long update cycle, and seriously affecting the security and reliability of AR glasses systems, the present application provides an augmented reality glasses system, glasses, and storage medium, which adopt the following technical solutions: In a first aspect, the present application provides an augmented reality glasses system, comprising: Iris feature extraction module, used to obtain iris texture features, detect iris liveness features, and obtain iris feature data; The heartbeat feature analysis module is used to collect the user's heart rate signal, extract the heart rate variability characteristics for liveness detection, and obtain heartbeat feature data; A liveness anti-counterfeiting module, configured to perform multi-dimensional liveness detection based on the iris feature data and the heartbeat feature data, evaluate the authenticity of the detection results, and obtain a liveness verification result; A dynamic key generation module, configured to generate an initial key based on the iris feature data and the heartbeat feature data, and dynamically update the initial key at a preset frequency to obtain a dynamic key; The data encryption module is used to encrypt user data based on the dynamic key.

[0006] By adopting the above technical solution, since traditional AR glasses use a single biometric feature for identity authentication and use static keys to encrypt data, they are vulnerable to counterfeiting attacks and key theft; for example, attackers can bypass iris recognition through high-definition photos or 3D printed models, or obtain fixed keys through long-term monitoring; this application first obtains the texture features and liveness features of the user's iris through the iris feature extraction module, while the heartbeat feature analysis module collects the heart rate signal and extracts its variability features; the live anti-counterfeiting module performs multi-dimensional fusion analysis on these two biometric data and verifies the user's identity by evaluating the authenticity of the features; then, the dynamic key generation module uses the verified iris feature data and heartbeat feature data to generate an initial encryption key, and periodically updates the key according to a preset frequency; finally, the data encryption module uses the dynamically updated key to encrypt the user data to ensure data transmission security; the anti-counterfeiting ability is improved through the collaborative verification of biometrics, and the encryption strength is enhanced by the real-time change of the key with the physiological characteristics, thereby improving the security and reliability of the AR glasses system.

[0007] Optionally, the iris feature extraction module specifically includes: The structured light projection unit is used to emit coded light spots to the iris area, collect reflected light information, and obtain depth information; a light spot analysis unit, configured to calculate light spot deformation characteristics according to the depth information to obtain surface relief data; a texture detection unit, configured to analyze the three-dimensional features of the iris texture according to the surface relief data to obtain a spatial feature map; A living body feature unit, configured to extract iris tissue features based on the spatial feature map, detect pupil reflections and blinking movements, and obtain living body feature indicators; The feature evaluation unit is used to perform authenticity scoring based on the living feature indicators to obtain an iris feature verification result.

[0008] By adopting the above technical solution, ordinary iris recognition systems are unable to effectively distinguish between real eyeballs and high-definition printed photos, and are prone to misjudgment. The structured light projection unit of the present application first emits a specifically coded structured light spot to the user's iris area, collects the reflected light information, and obtains the depth information of the iris surface through the principle of optical imaging. The light spot analysis unit then processes this depth information and calculates the deformation characteristics of the structured light spot on the iris surface, thereby obtaining microscopic undulation data of the iris surface. Based on this undulation data, the texture detection unit analyzes the three-dimensional structural characteristics of the iris texture and generates a feature map containing spatial information. The live feature unit further extracts the biological characteristics of the iris tissue from the spatial feature map, while monitoring the pupil's reaction to light and natural blinking movements to obtain live feature indicators. Finally, the feature evaluation unit performs a comprehensive score based on these three-dimensional feature indicators and outputs the verification result of the iris feature. Not only can richer iris feature information be obtained, but also forgeries can be effectively identified by analyzing the microscopic deformation patterns of the iris surface, while combining physiological reactions such as pupil reflection and blinking to determine liveness.

[0009] Optionally, the living anti-counterfeiting module specifically includes: a feature synchronization unit, configured to perform time sequence alignment on the iris feature data and the heartbeat feature data to obtain a synchronized feature sequence; A physiological characteristic analysis unit, configured to extract the iris-heartbeat coordinated variation rule based on the synchronous characteristic sequence to obtain a physiological characteristic pattern; an authenticity assessment unit, configured to calculate feature consistency and change continuity based on the physiological feature pattern to obtain a living body credibility; An attack detection unit is used to identify abnormal feature patterns based on the liveness credibility, determine whether there is a deception attack, and obtain an anti-counterfeiting detection result; The anti-counterfeiting decision unit is used to perform liveness verification according to the anti-counterfeiting detection result and trigger a security warning mechanism when a counterfeiting attack is detected.

[0010] By adopting the above-mentioned technical solution, the feature synchronization unit of the present application first performs time alignment processing on the collected iris feature data and heartbeat feature data to ensure the precise matching of the two feature data in the time dimension to form a synchronized feature sequence; the physiological feature analysis unit then analyzes these synchronized data, extracts the synergistic rules between the iris response and the heartbeat changes, and establishes the feature pattern under the normal physiological state; the authenticity assessment unit calculates the degree of consistency between the current collected data and the standard physiological pattern based on the feature pattern, and evaluates the continuity of the feature changes, thereby deriving the liveness credibility; the attack detection unit identifies possible abnormal feature patterns by comparing the liveness credibility with the preset security threshold, and outputs the anti-counterfeiting detection result; finally, the anti-counterfeiting decision unit determines whether there is a counterfeiting attack based on the detection result, and immediately triggers the security warning mechanism when an abnormality is found; it can effectively identify the respectively forged feature data, significantly improving the anti-counterfeiting ability of the system.

[0011] Optionally, the physiological characteristic analysis unit specifically includes: a feature matching subunit, configured to perform pattern matching on the synchronization feature sequence according to a preset iris-heartbeat physiological association model to obtain a feature matching degree; A time delay analysis subunit, configured to calculate a time delay parameter between an iris response and a heartbeat change according to the feature matching degree, and obtain physiological time delay data; A consistency verification subunit, configured to determine whether the physiological delay data conforms to a standard physiological response range and obtain a consistency verification result; a deviation calculation subunit, configured to quantify the degree of deviation between the current feature and the standard model according to the consistency verification result, and obtain the feature credibility; The threshold determination subunit is used to evaluate the authenticity of the living body based on the feature credibility, and determine it as an abnormal physiological response when the deviation exceeds a preset threshold.

[0012] By adopting the above-mentioned technical solution, the feature matching subunit of the present application first performs pattern matching analysis on the synchronous feature sequence based on the pre-established iris-heartbeat physiological association model, and calculates the degree of matching between the feature sequence and the standard model; the delay analysis subunit then accurately calculates the response delay of the iris feature to the heartbeat change based on the feature matching degree, and obtains the delay parameters reflecting the physiological coupling relationship; the consistency verification subunit compares these delay data with the normal physiological response range of the human body to evaluate the rationality of the feature change; the deviation calculation subunit further quantifies the deviation between the current feature and the standard physiological model, and generates a quantifiable feature credibility index; finally, the threshold judgment subunit evaluates the authenticity of the living body based on the feature credibility, and when it is found that the deviation exceeds the preset threshold, it is judged as an abnormal physiological response; by analyzing the temporal dependency between the iris and heartbeat features, a deep anti-counterfeiting mechanism based on physiological coupling delay is established.

[0013] Optionally, the dynamic key generation module specifically includes the following steps: a feature quantization unit, configured to digitally encode the iris feature data and the heartbeat feature data to obtain a feature quantization sequence; A feature entropy extraction unit, configured to calculate a biometric feature entropy value based on the feature quantization sequence and generate a random seed; A key generation unit, configured to generate an initial key using an elliptic curve encryption algorithm according to the random seed; A timestamp embedding unit, configured to embed timestamp information in the initial key and set a validity period for the key; A key updating unit, configured to periodically update the initial key according to the preset frequency and newly collected characteristic data to obtain a dynamic key; A key verification unit is used to perform integrity verification on the dynamic key.

[0014] By adopting the above technical solution, the existing technology generally adopts a fixed key generation algorithm and update cycle, lacks the full utilization of the randomness of biometrics, and results in insufficient key security and dynamics. The feature quantization unit of the present application first digitizes the verified iris feature data and heartbeat feature data to convert the biometrics into a computable feature quantization sequence. The feature entropy extraction unit then analyzes the information entropy characteristics of these quantization sequences and extracts the random components therein as the key seed. The key generation unit generates a high-strength initial key based on this bio-random seed using an elliptic curve encryption algorithm. The timestamp embedding unit injects time information into the initial key to set an expiration date for the key to ensure the timeliness of the key. The key update unit periodically updates the key according to the system's preset update frequency and in combination with newly collected biometric data to maintain the key's dynamic nature. Finally, the key verification unit performs integrity verification on the generated dynamic key to ensure the key's availability. The dynamic evolution of the key is achieved through periodic updates, and the timeliness of the key is ensured by combining the timestamp mechanism, forming a complete dynamic key management system.

[0015] Optionally, the feature entropy extraction unit specifically includes: A block processing subunit, configured to segment the feature quantization sequence into segments according to preset lengths to obtain feature data blocks; An information statistics subunit, configured to calculate the frequency of occurrence of different feature values ​​in each of the feature data blocks to obtain a feature distribution probability; An entropy value calculation subunit, configured to calculate the entropy value of each data block according to the feature distribution probability to obtain a local entropy value sequence; A weighted fusion subunit, configured to perform weighted summation on the local entropy value sequence to obtain a comprehensive entropy value; The random mapping subunit is used to convert the comprehensive entropy value into a key seed to obtain a random seed sequence.

[0016] By adopting the above technical solution, the block processing subunit first segments the feature quantization sequence according to a scientifically set length, dividing the continuous feature data into multiple feature data blocks to facilitate local entropy value analysis; the information statistics subunit then performs feature value statistics on each data block, calculates the frequency of occurrence of different feature values, and constructs a probability distribution model of the features; the entropy value calculation subunit uses the information entropy calculation formula based on these probability distribution data to evaluate the entropy value of each data block, and obtains an entropy value sequence that reflects local randomness; the weighted fusion subunit sets scientific weight coefficients to perform weighted combination of the entropy values ​​of each data block, and obtains a comprehensive entropy value that can fully reflect the randomness of the features; finally, the random mapping subunit maps the comprehensive entropy value to the key seed through nonlinear transformation, ensuring that the generated random seed sequence has a good randomness distribution; not only does it ensure the overall randomness of the random seed, but it also improves the uniformity of the randomness through local entropy value analysis, significantly enhancing the security of the key system.

[0017] Optionally, also include: A physiological status monitoring module is used to collect the heartbeat characteristic data and EEG and skin conductance signals, and combine them with the characteristic quantization sequence to obtain multi-dimensional physiological characteristic data; An emotion prediction module is used to perform time series analysis based on the multi-dimensional physiological feature data and the LSTM neural network, evaluate data reliability using the local entropy value sequence, and obtain emotion prediction results; A content adjustment module, configured to dynamically adjust the brightness, refresh rate, and interaction difficulty of the augmented reality display content based on the emotion prediction results to obtain optimized display parameters; A health management module is used to monitor the multi-dimensional physiological characteristic data in real time, trigger an early warning mechanism when the pressure exceeds a preset threshold, and generate a deep breathing guidance animation; The feedback optimization module is used to collect the user's response data to the guidance animation and adjust the warning threshold and guidance strategy according to the response data.

[0018] By adopting the above technical solution, the physiological status monitoring module first integrates heartbeat feature data with newly added EEG and skin conductance signals, and combines them with existing feature quantization sequences to construct a multi-dimensional physiological feature data system. The emotion prediction module then inputs these multi-dimensional features into the LSTM neural network for time series analysis, while innovatively using local entropy value sequences to evaluate the reliability of the data, thereby obtaining emotion prediction results with high confidence. The content adjustment module intelligently adjusts parameters such as the brightness and refresh rate of the AR display content based on the predicted emotional state, and appropriately reduces the difficulty of interaction to achieve dynamic optimization of the display effect. The health management module monitors the user's stress level through real-time analysis of multi-dimensional physiological feature data. When abnormal stress is detected, it promptly triggers an alert and generates a deep breathing guidance animation. The feedback optimization module continuously collects the user's response to the guidance animation and optimizes the warning threshold and guidance strategy based on the response data. This not only expands the application value of biometric data, but also improves the user experience through real-time adjustment and health intervention, giving the AR glasses system proactive health management capabilities.

[0019] In a second aspect, the present application provides a pair of glasses comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the module functions of the above-mentioned augmented reality glasses system when executing the computer program.

[0020] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the module functions of the above-mentioned augmented reality glasses system when executed by a processor.

[0021] In summary, this application includes at least one of the following beneficial technical effects: 1. This application uses an iris feature extraction module to obtain iris texture and liveness features, while a heartbeat feature analysis module collects heart rate signals and extracts variability features. A liveness anti-counterfeiting module performs a multi-dimensional fusion analysis of the two biometric features, assessing their authenticity to verify user identity. A dynamic key generation module uses verified feature data to generate an initial key and updates it at a preset frequency. A data encryption module uses a dynamic key to encrypt user data. This improves anti-counterfeiting capabilities through collaborative biometric verification, and enhances encryption strength by using physiological characteristics to drive dynamic key updates, thereby enhancing the security and reliability of the AR glasses system. 2. This application uses a structured light projection unit to emit a coded light spot toward the iris area and collect reflected light information to obtain iris surface depth information. A light spot analysis unit calculates the deformation characteristics of the structured light spot to obtain surface micro-undulation data. A texture detection unit analyzes the three-dimensional characteristics of the iris texture based on the undulation data to generate a spatial feature map. A live feature unit extracts iris tissue characteristics and monitors pupil reflections and blinking movements. A feature evaluation unit comprehensively scores the three-dimensional feature indicators and outputs a verification result. This system can obtain rich iris three-dimensional features and effectively identify counterfeit samples by analyzing surface micro-deformation and physiological reactions. 3. This application uses a feature matching subunit to perform matching analysis on the synchronous feature sequence based on the iris-heartbeat physiological correlation model; the delay analysis subunit calculates the response delay of the iris feature to the heartbeat change and obtains the physiological coupling relationship parameters; the consistency verification subunit compares the delay data with the normal physiological range to evaluate the rationality of the feature change; the deviation calculation subunit quantifies the deviation of the feature from the standard model to obtain the credibility index; the threshold judgment subunit evaluates the authenticity of the living body based on the credibility, and determines it as an abnormal response when an abnormality is found; by analyzing the temporal dependency between features, a physiological coupling delay anti-counterfeiting mechanism is established. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a schematic diagram of the structure of an augmented reality glasses system according to an embodiment of the present application. Figure 1 ; Figure 2 This is a structural diagram of an iris feature extraction module in an augmented reality glasses system according to an embodiment of the present application; Figure 3 This is a schematic structural diagram of a living anti-counterfeiting module in an augmented reality glasses system according to an embodiment of the present application; Figure 4 This is a schematic diagram of the structure of a physiological characteristic analysis unit in an augmented reality glasses system according to an embodiment of the present application; Figure 5 This is a schematic diagram of the structure of a dynamic key generation module in an augmented reality glasses system according to an embodiment of the present application; Figure 6 This is a schematic diagram of the structure of a feature entropy extraction unit in an augmented reality glasses system according to an embodiment of the present application; Figure 7 This is a schematic diagram of the structure of an augmented reality glasses system according to an embodiment of the present application. Figure 2 ; Figure 8 This is a diagram of the internal structure of a pair of glasses according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] The terms used in the following examples of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification and appended claims of this application, the singular expressions "a," "an," "said," "above," "the," and "this" are intended to include plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in this application refers to any or all possible combinations comprising one or more of the listed items.

[0024] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to imply or suggest relative importance or implicitly indicate the number of the technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of this application, unless otherwise specified, "plurality" means two or more.

[0025] The embodiments of the present application are described in further detail below with reference to the accompanying drawings.

[0026] In the first aspect, the present application provides an augmented reality glasses system, referring to Figure 1 ,include: The iris feature extraction module is used to obtain iris texture features, detect iris liveness features, and obtain iris feature data.

[0027] In this embodiment, iris features include surface texture features and deep tissue features. Surface texture features refer to the radial lines, annular wrinkles, pigment distribution, and crypt structures visible on the iris surface. Deep tissue features refer to the three-dimensional structural characteristics of the iris tissue, including tissue fiber distribution, blood vessel orientation, and muscle layers.

[0028] The heartbeat feature analysis module is used to collect the user's heart rate signal, extract the heart rate variability characteristics for liveness detection, and obtain heartbeat feature data.

[0029] In this embodiment, the heart rate signal refers to a physiological signal that reflects the rhythm of heartbeats, including heart rate frequency, heart beat intervals, and waveform characteristics. The heart rate variability characteristic refers to the dynamic changes in the heart beat intervals, reflecting the regulation of the autonomic nervous system on heart activity.

[0030] Specifically, the heartbeat feature analysis module integrates a photoplethysmography sensor on the inside of the temple of the glasses, with a sampling frequency of 200Hz, and continuously collects heart rate signals for 60 seconds. A wavelet transform is used to remove baseline drift and high-frequency noise, and extract the R-wave peak point sequence. The mean, standard deviation, and coefficient of variation of adjacent R-wave intervals are calculated as statistical features. The frequency domain energy distribution is analyzed to obtain the ratio of the low-frequency component (0.04-0.15Hz) to the high-frequency component (0.15-0.4Hz). The characteristic vectors of 10 sets of statistical parameters are established to characterize heart rate variability.

[0031] The liveness anti-counterfeiting module is used to perform multi-dimensional liveness detection based on iris feature data and heartbeat feature data, evaluate the authenticity of the detection results, and obtain liveness verification results.

[0032] Specifically, the live anti-counterfeiting module first independently verifies the authenticity of the iris features and heart rate features. The iris feature matching threshold is set to 0.85, and the number of feature point matches is not less than 80. The heart rate parameter requires the heart rate to be in the range of 45-120 beats / minute and the coefficient of variation to be between 0.02-0.08. After both features pass the preliminary verification, the dynamic change pattern of the features is further analyzed. A feature mapping table is established to record the correspondence between iris response and heart rate changes under normal physiological conditions, which is used to detect the synergy of feature changes. When it is detected that the feature changes do not conform to normal physiological laws, it is determined that a counterfeit attack may have occurred.

[0033] The dynamic key generation module is used to generate an initial key based on iris feature data and heartbeat feature data, and dynamically update it at a preset frequency to obtain a dynamic key.

[0034] In this embodiment, the initial key refers to the basic encryption key generated after the system passes initial authentication. Its generation process incorporates the randomness of iris and heart rate biometrics. Dynamic update refers to the periodic refreshing of key content based on real-time changes in biometrics. The preset frequency refers to the interval between key updates, ensuring key timeliness.

[0035] Specifically, the dynamic key generation module first digitally encodes the verified iris feature matrix and heart rate feature sequence. The iris feature matrix is ​​quantized to 8 bits, and the pixel values ​​of the middle 128×128 area are selected as the basic sequence. The heart rate feature sequence is normalized, and the 4 digits after the decimal point are taken to construct a 32-bit supplementary sequence. The two sequences are XORed to obtain a mixed sequence, and a 256-bit initial key is generated using the SHA-256 hash algorithm. The system collects new biometric data every 10 minutes, uses the same method to generate an update sequence, and performs a permutation operation with the current key to achieve dynamic key updates.

[0036] The data encryption module is used to encrypt user data based on a dynamic key.

[0037] In this embodiment, data encryption refers to the process of protectively transforming user data using a key. User data includes four categories of information: device configuration information, personal identification information, usage records, and interaction data. Encryption processing involves converting plaintext data into ciphertext using a specific encryption algorithm to ensure data security during transmission and storage.

[0038] In one embodiment, referring to Figure 2 , the iris feature extraction module specifically includes: The structured light projection unit is used to emit coded light spots to the iris area, collect reflected light information, and obtain depth information.

[0039] In this embodiment, the coded light spot refers to a structured light pattern with a specific spatial distribution pattern, including three basic patterns: dot matrix, stripe, and grid. Reflected light information refers to the spatial distribution of the deformation of the light spot after it is projected onto the iris surface. Depth information refers to the three-dimensional contour data of the iris surface obtained through optical triangulation.

[0040] Specifically, the structured light projection unit uses a near-infrared wavelength semiconductor laser array to generate a dotted light spot. Precision-designed diffractive optical elements project the dotted light spot evenly onto the iris area, with the spot spacing and diameter optimized based on iris size. The system pre-establishes a standard planar light spot database, storing the coordinate matrix of the light spot position and light intensity distribution under a baseline state. A high-speed CMOS image sensor is used to capture reflected light information, creating a phase-depth mapping table. Phase offset values ​​are quickly converted into depth data through a table lookup.

[0041] The light spot analysis unit is used to calculate the light spot deformation characteristics according to the depth information to obtain surface relief data.

[0042] In this embodiment, the spot deformation characteristics refer to the characteristic parameters of the structured spot on the iris surface in three dimensions: displacement, deformation, and intensity change. The surface relief data refers to the microscopic height relief distribution map of the iris surface, reflecting the three-dimensional structural characteristics of the iris tissue.

[0043] Specifically, the spot analysis unit establishes a spot deformation feature mapping table, recording the standard values ​​of characteristic parameters under 10 basic deformation modes. First, the centroid coordinates, area, and grayscale distribution of each spot are extracted and compared with the reference pattern to calculate the displacement vector and deformation coefficient. The spot deformation field is fitted using the least squares method to establish a surface normal vector distribution map. The surface height distribution is reconstructed through integral operations to generate a micro-relief data matrix.

[0044] The texture detection unit is used to analyze the three-dimensional features of the iris texture according to the surface relief data to obtain a spatial feature map.

[0045] In this embodiment, the iris texture three-dimensional features refer to the spatial geometric features of the iris surface texture, including wrinkle depth, texture direction and surface roughness. The spatial feature map refers to the iris texture feature description containing depth information, which is used to characterize the three-dimensional structural characteristics of the iris tissue.

[0046] The living feature unit is used to extract iris tissue features based on the spatial feature map, detect pupil reflection and blinking movements, and obtain living feature indicators.

[0047] In this embodiment, iris tissue characteristics refer to histological features that reflect the biological activity of the iris, including tissue fiber distribution, blood vessel orientation, and pigment distribution. Pupillary reflexes and blinking movements refer to the natural physiological reactions of the eye to external stimuli and are used to verify the liveness of the iris.

[0048] Specifically, the spatial arrangement characteristics of tissue fibers are identified by analyzing the depth gradient distribution in the spatial feature map. A light intensity threshold detection method is used to monitor pupil contraction in response to light, and natural blinking is detected by tracking eyelid movement.

[0049] The feature evaluation unit is used to perform authenticity scoring based on the live feature indicators to obtain iris feature verification results.

[0050] In this embodiment, the authenticity scoring refers to a quantitative evaluation of the authenticity of the iris by comprehensively analyzing the three-dimensional structure and physiological response of the iris features. The iris feature verification result refers to the determination of the authenticity of the iris under test based on the scoring result.

[0051] In one embodiment, referring to Figure 3 The live anti-counterfeiting module specifically includes: The feature synchronization unit is used to perform time sequence alignment on the iris feature data and the heartbeat feature data to obtain a synchronized feature sequence.

[0052] In this embodiment, time alignment refers to the process of calibrating and matching iris features and heartbeat features with different sampling frequencies in the time dimension. The synchronized feature sequence refers to the feature data stream after time alignment, which contains the corresponding time point data of the iris feature values ​​and heartbeat feature values.

[0053] The physiological characteristic analysis unit is used to extract the iris-heartbeat coordinated change law based on the synchronous characteristic sequence to obtain the physiological characteristic pattern.

[0054] In this embodiment, the iris-heartbeat synergy refers to the physiological coupling relationship between the two biometric features, primarily manifested in the amplitude, frequency, and phase of the feature changes. Physiological feature patterns refer to patterns of feature changes that conform to the natural physiological laws of the human body and are used to distinguish real biometric features from artificially simulated ones.

[0055] Specifically, the physiological feature analysis unit pre-establishes a physiological association rule base, recording the correspondence between iris responses and heart rate changes under typical conditions. A sliding window method is used to calculate the correlation coefficient of the feature sequence and extract a synergistic index of feature changes. A feature extraction table based on frequency domain analysis is established to identify periodic variation components in the feature sequence. The extracted variation patterns are then matched against the rule base to determine whether the feature changes conform to physiological laws.

[0056] The authenticity assessment unit is used to calculate the feature consistency and change continuity based on the physiological feature pattern to obtain the liveness credibility.

[0057] Specifically, the authenticity assessment unit establishes a scoring standard table encompassing three evaluation dimensions: feature similarity, change trend, and mutation amplitude. Consistency is assessed by calculating the deviation between the feature sequence and the standard pattern. A sliding variance method is used to detect mutation points in the sequence and assess the continuity of change. Each evaluation indicator is weighted and combined to generate a comprehensive credibility score.

[0058] The attack detection unit is used to identify abnormal feature patterns based on the liveness credibility, determine whether there is a deception attack, and obtain anti-counterfeiting detection results.

[0059] In this embodiment, an abnormal feature pattern refers to a pattern of feature changes that deviate from normal physiological patterns, including sudden changes in features, irregular cycles, and broken associations. A spoofing attack refers to the illegal use of artificial means to simulate biometric features. The anti-counterfeiting detection result refers to the risk assessment result of the detected abnormal features.

[0060] The anti-counterfeiting decision unit is used to perform liveness verification based on the anti-counterfeiting detection results and trigger a security warning mechanism when a counterfeiting attack is detected.

[0061] In one embodiment, referring to Figure 4 , the physiological characteristics analysis unit specifically includes: The feature matching subunit is used to perform pattern matching on the synchronous feature sequence according to a preset iris-heartbeat physiological association model to obtain a feature matching degree.

[0062] Specifically, the feature matching subunit pre-creates a physiological feature correspondence table, recording the standard variation patterns of iris features at different heart rate levels. A sliding window method is used to segment the feature sequence and calculate the feature correlation within each window. A table lookup is used to quickly obtain the standard iris response pattern corresponding to the current heart rate. The measured sequence is then compared with the standard pattern to determine the feature match.

[0063] The time delay analysis subunit is used to calculate the time delay parameters between the iris response and the heartbeat change according to the feature matching degree to obtain the physiological time delay data.

[0064] In this embodiment, iris response refers to the dynamic changes in iris characteristics as heart rate changes. The delay parameter refers to the lag time between iris characteristic changes and heart rate changes, reflecting the response characteristics of the physiological system. Physiological delay data refers to the time series that describes the delay pattern of characteristic responses.

[0065] Specifically, the delay analysis subunit establishes a delay parameter mapping table to store the characteristic response delay range under normal physiological conditions. It uses peak detection to identify the inflection point of heart rate changes, records the time when the iris feature reaches the corresponding state, and calculates the time difference as the delay parameter.

[0066] The consistency verification subunit is used to determine whether the physiological delay data meets the standard physiological response range and obtain a consistency verification result.

[0067] Specifically, the consistency verification subunit establishes a response range determination table, recording the normal intervals for delay parameters under multiple physiological conditions. It uses an interval determination method to assess the rationality of the delay data, and determines consistency when the parameters fall within the normal range. By combining multiple consecutive determination results, the reliability of the verification is improved.

[0068] The deviation calculation subunit is used to quantify the degree of deviation between the current feature and the standard model based on the consistency verification result to obtain the feature credibility.

[0069] Specifically, the deviation calculation subunit establishes a deviation scoring rule table, setting corresponding scoring criteria for different types of feature deviations. The degree of deviation is measured by calculating the distance between the feature sequence and the standard sequence. A piecewise linear mapping method is used to convert the deviation value into a credibility score to ensure the interpretability of the scoring results.

[0070] Furthermore, feature evaluation is performed from five dimensions: amplitude deviation, timing deviation, persistence deviation, correlation deviation, and waveform deviation. Amplitude deviation is divided into strong changes, medium changes, and weak changes, and the response amplitude of the iris feature is evaluated based on the degree of heart rate change; timing deviation includes precursor changes, synchronous changes, and lagging changes, and determines the temporal relationship between feature changes and heart rate changes; persistence deviation distinguishes continuous changes, intermittent changes, and sudden changes, and evaluates the smoothness of feature changes; correlation deviation is divided into strong correlation, medium correlation, and weak correlation, and quantifies the correlation between feature sequences and heart rate sequences; waveform deviation includes periodic changes, quasi-periodic changes, and irregular changes, and analyzes the regularity of feature changes. A feature deviation quantitative index table is established, each dimension is scored, and the final feature credibility is obtained through weighted combination, providing a basis for subsequent threshold determination.

[0071] The threshold judgment subunit is used to evaluate the authenticity of the living body based on the feature credibility, and when the deviation exceeds the preset threshold, it is judged as an abnormal physiological response.

[0072] In one embodiment, referring to Figure 5 , the dynamic key generation module specifically includes the following steps: The feature quantization unit is used to digitally encode the iris feature data and the heartbeat feature data to obtain a feature quantization sequence.

[0073] The feature entropy extraction unit is used to calculate the biometric feature entropy value based on the feature quantization sequence and generate a random seed.

[0074] In this embodiment, the biometric entropy value refers to the amount of random information contained in the feature data, reflecting the uncertainty of the feature sequence. The random seed refers to the random value used to initialize the key generation algorithm. Its randomness directly affects the security strength of the key.

[0075] Specifically, the feature entropy extraction unit calculates the information entropy of the feature sequence using a block statistical method, establishes an entropy calculation rule table, and records the entropy contributions corresponding to different feature patterns.

[0076] The key generation unit is used to generate an initial key according to a random seed using an elliptic curve encryption algorithm.

[0077] In this embodiment, the elliptic curve cryptography algorithm refers to an asymmetric encryption algorithm based on the elliptic curve discrete logarithm problem. The initial key refers to the basic key generated for the first time, which serves as the starting point for subsequent dynamic updates.

[0078] Specifically, the key generation unit pre-stores an elliptic curve parameter table and selects curve parameters with high security. The key generation process is initialized with a random seed, and a point multiplication operation is used to generate a public-private key pair. A key derivation function is established to derive the actual encryption key from the private key.

[0079] The timestamp embedding unit is used to embed timestamp information in the initial key and set the key validity period.

[0080] In this embodiment, the timestamp information refers to the time stamp of the key generation moment, which is used to track the timeliness of the key. The key validity period refers to the maximum time that a single key can be used. The key needs to be updated after the time exceeds this period.

[0081] The key updating unit is used to periodically update the initial key according to the preset frequency and the newly collected characteristic data to obtain a dynamic key.

[0082] In this embodiment, the preset frequency refers to the time interval between key updates, which affects the balance between system security and performance. A dynamic key refers to an encryption key that maintains timeliness through a periodic update mechanism.

[0083] Specifically, the key update unit establishes an update trigger rule table, determines the update timing based on the time interval and the magnitude of the feature change, and uses a key derivation function to generate an update sequence from the new feature data, which is mixed with the current key to obtain a new key.

[0084] The key verification unit is used to perform integrity verification on the dynamic key.

[0085] In this embodiment, integrity checking refers to verifying whether the key is tampered with or damaged during the generation and use process.

[0086] In one embodiment, referring to Figure 6 , the feature entropy extraction unit specifically includes: The block processing subunit is used to segment the feature quantization sequence according to a preset length to obtain feature data blocks.

[0087] In this embodiment, the feature quantization sequence refers to a digitally encoded biometric binary stream. The feature data block refers to dividing a continuous feature sequence into data segments of fixed length to facilitate local statistical analysis.

[0088] Specifically, a block parameter table is established to record the optimal block length corresponding to different feature types. A sliding block method is used to process feature sequences, and a certain overlap area is retained between adjacent data blocks to improve the continuity of feature extraction.

[0089] The information statistics subunit is used to calculate the occurrence frequency of different feature values ​​in each feature data block to obtain the feature distribution probability.

[0090] In this embodiment, the characteristic value refers to the different binary combinations that appear in the data block. The characteristic distribution probability refers to the normalized frequency of occurrence of each characteristic value in the data block, reflecting the statistical characteristics of the feature.

[0091] Specifically, a feature count table is established to record the number of occurrences of different feature values ​​in each data block. Histogram statistics are used to quickly calculate feature distribution, and probability normalization is completed through table lookup.

[0092] The entropy value calculation subunit is used to calculate the entropy value of each data block according to the feature distribution probability to obtain a local entropy value sequence.

[0093] In this embodiment, the local entropy value refers to the information entropy of a single data block, reflecting the degree of randomness of the data block. The local entropy value sequence refers to the time series composed of the entropy values ​​of all data blocks, reflecting the dynamic changes in the characteristic randomness.

[0094] Specifically, the entropy calculation subunit establishes an entropy calculation rule table and uses the information entropy calculation formula. It obtains the entropy contribution corresponding to the probability through table lookup and accumulates it to obtain the total entropy value of the data block. It also establishes an entropy normalization table and maps the calculation results to a standard range.

[0095] Furthermore, a basic entropy lookup table is established, dividing the probability value interval into a low-probability interval corresponding to slight changes in iris texture, an intermediate-probability interval corresponding to normal dynamic changes in the iris, and a high-probability interval corresponding to drastic pupil contraction; a feature weight table is constructed, dividing the iris data block into a core area around the pupil and an iris peripheral edge area. The core area around the pupil is assigned a higher weight coefficient because it directly responds to autonomic nervous system activity, while the peripheral edge area is assigned a lower weight coefficient as an auxiliary verification area; a feature combination entropy table is established to record typical patterns of iris-heartbeat coordinated changes, identify the positive correlation pattern of pupil dilation when the heart rate accelerates and reduce the entropy value to reflect its regularity, and identify the non-correlated pattern of random iris micro-movement when the heart rate is stable and increase the entropy value to indicate its randomness; a dynamic calibration mechanism is introduced to adjust the entropy calculation parameters according to the heart rate change trend, increase the calibration factor in the stage of drastic heart rate fluctuation to adapt to rapid response, and reduce the calibration factor in the stage of stable heart rate to maintain stable output; finally, the entropy value result reflecting the authenticity of the physiological characteristics is comprehensively calculated through the entropy combination strategy table.

[0096] The weighted fusion subunit is used to perform weighted summation on the local entropy value sequence to obtain the comprehensive entropy value.

[0097] In this embodiment, weighted summation refers to weighted combination of local entropy values ​​according to the importance of different data blocks. The comprehensive entropy value is an overall measure of the randomness of the entire feature sequence and is used to generate high-quality random seeds.

[0098] Specifically, the weighted fusion subunit establishes a weight distribution table and determines the weight coefficient according to the location and content characteristics of the data block. The local entropy value is fused using piecewise linear combination. The random mapping subunit is used to convert the comprehensive entropy value into a key seed to obtain a random seed sequence.

[0099] In this embodiment, the key seed refers to a random value used to initialize the key generation process. The random seed sequence refers to a random bit stream converted from the comprehensive entropy value, which directly determines the randomness of the generated key.

[0100] In one embodiment, referring to Figure 7 , also includes: The physiological status monitoring module is used to collect heartbeat characteristic data and EEG and skin conductance signals, and combine them with the characteristic quantization sequence to obtain multi-dimensional physiological characteristic data.

[0101] Specifically, a signal acquisition parameter table is established to specify the sampling frequency, signal bandwidth, and filtering parameters for each sensor type. Time-division multiplexing is used to coordinate multi-channel signal acquisition, and signal preprocessing is rapidly completed through table lookup. A feature fusion rule table is established to time-align different types of physiological signals with the quantized iris feature sequence. A feature correlation mapping table is constructed to record the correspondence between iris features and other physiological signals for verifying signal validity. A data quality assessment table is established to weight other physiological signals based on the stability of the iris feature sequence and eliminate unreliable data segments. A layered data packet format is used, with iris features as the baseline time series. Other physiological features are then packaged according to timestamp alignment to generate a feature data packet in a standard format.

[0102] The emotion prediction module is used to perform time series analysis based on multi-dimensional physiological feature data and LSTM neural network, and use the local entropy value sequence to evaluate data reliability to obtain emotion prediction results.

[0103] In this embodiment, the emotion prediction result includes emotion category, intensity score and credibility index, and the prediction of the user's emotional state is achieved through time series analysis.

[0104] Specifically, an LSTM model is pre-established, retaining only the core gating units to reduce computational complexity. A feature map is created to convert the raw physiological data into a network input format. Prediction results are weighted using a sequence of local entropy values, eliminating low-confidence predictions. High local entropy values ​​indicate that the physiological characteristics of that period are highly random and uncertain, resulting in low-confidence predictions. Moderate local entropy values ​​indicate that the feature data is stable and contains sufficient information, resulting in high-confidence predictions. Low local entropy values ​​indicate that the feature data may be saturated or degraded, also leading to decreased prediction confidence. First, a sliding average is performed on the local entropy sequence to produce a smoothed entropy curve. This entropy curve is then mapped to corresponding weight coefficients, and the emotion predictions output by the LSTM model are weighted frame by frame to produce a weighted prediction sequence. A confidence threshold is set. When the weight of a frame's prediction falls below the threshold, the predictions of adjacent frames are interpolated to ensure continuity of the output sequence.

[0105] The content adjustment module is used to dynamically adjust the brightness, refresh rate and interaction difficulty of the augmented reality display content according to the emotion prediction results to obtain optimized display parameters.

[0106] In this embodiment, optimizing display parameters refers to dynamically adjusting display settings according to the user's emotional state, including screen brightness, refresh rate, and interaction complexity.

[0107] The health management module is used to monitor multi-dimensional physiological characteristic data in real time. When the pressure exceeds the preset threshold, the early warning mechanism is triggered and a deep breathing guidance animation is generated.

[0108] In this embodiment, the early warning mechanism refers to an intervention process initiated when an abnormal stress level is detected in the user. The deep breathing guidance animation refers to interactive animation content that guides the user to regulate breathing through visual rhythms, helping the user maintain a healthy state through active intervention.

[0109] Specifically, the health management module establishes a stress assessment rule table and calculates the stress index based on multi-dimensional physiological characteristics. It uses a sliding window method to monitor the pressure change trend and triggers an early warning based on the threshold judgment. It also establishes an animation template library to store breathing guidance animations of different rhythms and select the appropriate guidance mode according to the user's status. The feedback optimization module is used to collect user response data to the guidance animation and adjust the warning threshold and guidance strategy based on the response data.

[0110] In this embodiment, the response data refers to the changes in the user's physiological indicators when the user performs the guidance action. The guidance strategy refers to the combination configuration of the warning trigger condition and the guidance animation parameters.

[0111] Specifically, the feedback optimization module establishes a response evaluation table to record the degree of improvement in physiological indicators during the guidance process. Statistical methods are used to analyze the response effects and adjust warning thresholds and guidance parameters. A user preference record table is also established to store personalized guidance strategy configurations.

[0112] In one embodiment, the present application provides a pair of glasses, the internal structure of which can be as follows: Figure 8 As shown. The glasses include a processor, memory, and a network interface connected via a system bus. The processor of the glasses is used to provide computing and control capabilities. The memory of the glasses includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the glasses is used to store data. The network interface of the glasses is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, it realizes the module functions of an augmented reality glasses system.

[0113] Those skilled in the art will understand that Figure 8 The structure shown in the figure is merely a block diagram of a portion of the structure related to the present application solution, and does not constitute a limitation on the glasses to which the present application solution is applied. Specific glasses may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0114] In one embodiment, glasses are further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0115] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The above-described computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media used in the embodiments provided herein may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0116] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. An augmented reality glasses system, characterized in that: include: Iris feature extraction module, used to obtain iris texture features, detect iris liveness features, and obtain iris feature data; The heartbeat feature analysis module is used to collect the user's heart rate signal, extract the heart rate variability characteristics for liveness detection, and obtain heartbeat feature data; A liveness anti-counterfeiting module, configured to perform multi-dimensional liveness detection based on the iris feature data and the heartbeat feature data, evaluate the authenticity of the detection results, and obtain a liveness verification result; A dynamic key generation module, configured to generate an initial key based on the iris feature data and the heartbeat feature data, and dynamically update the initial key at a preset frequency to obtain a dynamic key; The data encryption module is used to encrypt user data based on the dynamic key.

2. The augmented reality glasses system according to claim 1, wherein: The iris feature extraction module specifically includes: The structured light projection unit is used to emit coded light spots to the iris area, collect reflected light information, and obtain depth information; a light spot analysis unit, configured to calculate light spot deformation characteristics according to the depth information to obtain surface relief data; a texture detection unit, configured to analyze the three-dimensional features of the iris texture according to the surface relief data to obtain a spatial feature map; A living body feature unit, configured to extract iris tissue features based on the spatial feature map, detect pupil reflections and blinking movements, and obtain living body feature indicators; The feature evaluation unit is used to perform authenticity scoring based on the living feature indicators to obtain an iris feature verification result.

3. The augmented reality glasses system according to claim 1, wherein: The living anti-counterfeiting module specifically includes: a feature synchronization unit, configured to perform time sequence alignment on the iris feature data and the heartbeat feature data to obtain a synchronized feature sequence; A physiological characteristic analysis unit, configured to extract the iris-heartbeat coordinated variation rule based on the synchronous characteristic sequence to obtain a physiological characteristic pattern; an authenticity assessment unit, configured to calculate feature consistency and change continuity based on the physiological feature pattern to obtain a living body credibility; An attack detection unit is used to identify abnormal feature patterns based on the liveness credibility, determine whether there is a deception attack, and obtain an anti-counterfeiting detection result; The anti-counterfeiting decision unit is used to perform liveness verification according to the anti-counterfeiting detection result and trigger a security warning mechanism when a counterfeiting attack is detected.

4. The augmented reality glasses system according to claim 3, wherein: The physiological characteristic analysis unit specifically includes: a feature matching subunit, configured to perform pattern matching on the synchronization feature sequence according to a preset iris-heartbeat physiological association model to obtain a feature matching degree; A time delay analysis subunit, configured to calculate a time delay parameter between an iris response and a heartbeat change according to the feature matching degree, and obtain physiological time delay data; A consistency verification subunit, configured to determine whether the physiological delay data conforms to a standard physiological response range and obtain a consistency verification result; a deviation calculation subunit, configured to quantify the degree of deviation between the current feature and the standard model according to the consistency verification result, and obtain the feature credibility; The threshold determination subunit is used to evaluate the authenticity of the living body based on the feature credibility, and determine it as an abnormal physiological response when the deviation exceeds a preset threshold.

5. The augmented reality glasses system according to claim 1, wherein: The dynamic key generation module specifically includes the following steps: a feature quantization unit, configured to digitally encode the iris feature data and the heartbeat feature data to obtain a feature quantization sequence; A feature entropy extraction unit, configured to calculate a biometric feature entropy value based on the feature quantization sequence and generate a random seed; A key generation unit, configured to generate an initial key using an elliptic curve encryption algorithm according to the random seed; A timestamp embedding unit, configured to embed timestamp information in the initial key and set a validity period for the key; A key updating unit, configured to periodically update the initial key according to the preset frequency and newly collected characteristic data to obtain a dynamic key; A key verification unit is used to perform integrity verification on the dynamic key.

6. The augmented reality glasses system according to claim 5, characterized in that: The feature entropy extraction unit specifically includes: A block processing subunit, configured to segment the feature quantization sequence into segments according to a preset length to obtain feature data blocks; An information statistics subunit, configured to calculate the frequency of occurrence of different feature values ​​in each of the feature data blocks to obtain a feature distribution probability; An entropy value calculation subunit, configured to calculate the entropy value of each data block according to the feature distribution probability to obtain a local entropy value sequence; A weighted fusion subunit, configured to perform weighted summation on the local entropy value sequence to obtain a comprehensive entropy value; The random mapping subunit is used to convert the comprehensive entropy value into a key seed to obtain a random seed sequence.

7. The augmented reality glasses system according to claim 6, characterized in that: Also includes: A physiological status monitoring module is used to collect the heartbeat characteristic data and EEG and skin conductance signals, and combine them with the characteristic quantization sequence to obtain multi-dimensional physiological characteristic data; An emotion prediction module is used to perform time series analysis based on the multi-dimensional physiological feature data and the LSTM neural network, evaluate data reliability using the local entropy value sequence, and obtain emotion prediction results; A content adjustment module, configured to dynamically adjust the brightness, refresh rate, and interaction difficulty of the augmented reality display content based on the emotion prediction results to obtain optimized display parameters; A health management module is used to monitor the multi-dimensional physiological characteristic data in real time, trigger an early warning mechanism when the pressure exceeds a preset threshold, and generate a deep breathing guidance animation; The feedback optimization module is used to collect the user's response data to the guidance animation and adjust the warning threshold and guidance strategy according to the response data.

8. A pair of glasses, characterized in that: The system comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the module functions of the augmented reality glasses system according to any one of claims 1 to 7 when executing the computer program.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the module functions of the augmented reality glasses system according to any one of claims 1 to 7 are realized.

Images