Intelligent signboard verification management method based on biological characteristics

Abstract

The invention relates to the technical field of biological feature recognition, and discloses a biological feature-based smart signboard verification management method, which comprises the following steps of: establishing a living body recognition model based on historical living body feature data; generating a living body recognition result H3 of the current user in combination with the obtained living body feature data of the current user and a living body recognition model; establishing a biological feature information base based on the fingerprint feature information and the facial feature information of all the users; retrieving a biological characteristic information base based on the obtained biological characteristic information of the current user, and generating an identity recognition result F of the current user; and generating a signboard verification result in combination with the identity recognition result F of the current user and the living body recognition result H3. Accurate living body recognition can be carried out according to comprehensive features of facial actions and illumination, and the adaptability of the model in various practical application scenes is improved.

Description

Technical Field

[0001] The present invention relates to the field of biometric identification technology, and in particular to a biometric-based smart sign verification and management method. Background Art

[0002] With the development of the Internet of Things and smart home, biometric recognition technology has also been widely used in various devices and systems, such as smart door locks, surveillance cameras, access control systems, etc. The needs of these application scenarios have driven the continuous innovation and development of biometric recognition technology.

[0003] In smart signage verification management methods, the application of biometric recognition technology can improve the accuracy and security of signage verification. Liveness detection is a key issue in biometric recognition. Attackers can use fake biometrics (such as photos, videos, or 3D models) to deceive the recognition system. Therefore, accurately detecting the liveness of biometrics is a challenge. Summary of the Invention

[0004] The purpose of the present invention is to prevent attacks that forge biometric features by combining a liveness recognition model with a biometric information database, monitoring data such as temperature, pulse, capacitance, and judging facial movements and lighting; a fast and accurate identity authentication process reduces user waiting time and improves user experience.

[0005] To achieve the above objectives, the present invention provides a biometric-based smart sign verification and management method, comprising:

[0006] Establish a liveness recognition model based on historical liveness feature data;

[0007] Combining the acquired live feature data of the current user with the live recognition model to generate the live recognition result H3 of the current user;

[0008] Enter the fingerprint feature information and facial feature information of all users to establish a biometric information database; search the biometric information database based on the acquired fingerprint feature information and facial feature information of the current user to generate the identity recognition result F of the current user;

[0009] Combine the current user's identity recognition result F and the liveness recognition result H3 to generate the identification card verification result;

[0010] The historical living body characteristic data includes: historical first living body characteristic data and historical second living body characteristic data.

[0011] In some embodiments of the present invention, generating the liveness recognition result H3 of the current user includes:

[0012] The first historical living feature data includes: historical temperature data, historical pulse data, and historical capacitance data;

[0013] Establishing a fingerprint liveness recognition model based on the first liveness feature data, and generating a first-level liveness recognition result H1 by combining the fingerprint liveness recognition model and the acquired real-time first liveness feature data;

[0014] The historical second living feature data includes: historical facial illumination data and historical facial movement data;

[0015] Establishing a facial liveness recognition model based on the second liveness feature data, and generating a secondary liveness recognition result H2 by combining the established facial liveness recognition model and the acquired real-time second liveness feature data;

[0016] Generate a liveness recognition result H3 of the current user based on the first-level liveness recognition result H1 and the second-level liveness recognition result H2;

[0017] When H1=1 and H2=1, H3=1; when H1≠1 or H2≠1, H3=0.

[0018] In some embodiments of the present invention, the establishment of a fingerprint liveness recognition model includes:

[0019] Historical temperature data, historical pulse data, historical capacitance data;

[0020] Generate a first living body temperature interval (a1, a2) based on the preset temperature interval of historical temperature data;

[0021] Generate a temperature interval correction coefficient b based on the current ambient temperature, and generate a second living body temperature interval (a1*(1-b), a2*(1+b)) by combining the temperature interval correction coefficient and the first living body temperature interval;

[0022] Set pulse monitoring intervals (c1, c2) based on historical pulse data;

[0023] Set the capacitance monitoring interval (d1, d2) based on historical capacitance data;

[0024] Wherein, a1 represents a first preset value of living body temperature, a2 represents a second preset value of living body temperature, c1 represents a first preset value of pulse monitoring, c2 represents a second preset value of pulse monitoring, d1 represents a first preset value of capacitance monitoring, and d2 represents a second preset value of capacitance monitoring;

[0025] A fingerprint liveness recognition model is generated by combining the second living body temperature interval (a1*(1-b), a2*(1+b)), the pulse monitoring interval (c1, c2) and the capacitance monitoring interval (d1, d2).

[0026] In some embodiments of the present invention, generating the first-level liveness recognition result H1 includes:

[0027] A temperature sensor, a light emitting diode, a photodetector and an AC electric field are arranged on the fingerprint sensor;

[0028] Generate user's real-time temperature data AI based on temperature sensor;

[0029] Generate user's real-time pulse data ci based on light emitting diodes and photodetectors;

[0030] Based on the AC electric field, the electric field change data when the user uses the fingerprint sensor is monitored, and the user's real-time capacitance data di is generated based on the electric field change data;

[0031] Generate real-time first living body characteristic data, including: user's real-time temperature data ai, user's real-time pulse data ci, and user's real-time capacitance data di;

[0032] Generate a first-level liveness recognition result H1 based on the generated real-time first liveness feature data;

[0033] If ai∈(a1*(1-b), a2*(1+b))) and (ci∈(c1, c2)) and (di∈(d1, d2), then H1=1;

[0034] If ai∈(1-(a1*(1-b), a2*(1+b))) or ci∈1-(c1, c2) or di∈1-(d1, d2), then H1=0.

[0035] In some embodiments of the present invention, the method of generating the user's real-time pulse data ci based on the light emitting diode and the photodetector includes:

[0036] The light emitted by the light emitting diode and transmitted through the finger is monitored by a photodetector;

[0037] The light intensity of the light changes periodically with the pulse beat, and the real-time pulse data ci of the user is generated.

[0038] In some embodiments of the present invention, establishing a facial liveness recognition model based on the second liveness feature data includes:

[0039] extracting a facial feature dataset based on historical facial illumination data and historical facial motion data, wherein the facial feature dataset includes state information of a plurality of data points;

[0040] The state information of the data point includes the location information of the data point and the illumination information of the data point;

[0041] Acquire monitoring data of position information of a plurality of data points corresponding to historical facial action data to generate a facial action judgment model;

[0042] Acquire monitoring data of illumination information of a plurality of data points corresponding to historical facial illumination data to generate a facial illumination judgment model;

[0043] A facial liveness recognition model is established based on the facial action judgment model and the facial lighting judgment model.

[0044] In some embodiments of the present invention, generating a facial illumination determination model includes:

[0045] Conduct lighting experiments based on the acquired historical facial lighting data;

[0046] Obtain the light reflection intensity of the test target facial data points under different colors of light;

[0047] Calculate the variance of the light reflection intensity of all facial data points of the test target under the current color illumination, and generate a light reflection intensity consistency evaluation value based on the variance of the light reflection intensity of all facial data points;

[0048] Calculate the light reflection intensity consistency evaluation value corresponding to all color lighting, and compare it with the preset value to generate the facial lighting judgment result;

[0049] A facial lighting judgment model is generated by integrating the lighting color type and the facial lighting judgment results.

[0050] In some embodiments of the present invention, generating the secondary living body recognition result H2 includes:

[0051] Set the facial action in the facial recognition process and obtain the location information set E of the data points corresponding to the facial action performed by the current user, where E = {e1, e2...ei...en};

[0052] Among them, ei represents the location information of the i-th data point, and n represents the total number of data points;

[0053] The position information set E of the data points is sampled multiple times to obtain the position change data of the data points corresponding to the facial action performed by the current user;

[0054] Combining the position change data of the corresponding data points and the facial action judgment model to generate the facial action liveness judgment result M1;

[0055] If the position change data of the data point corresponding to the facial action performed by the current user obtained through multiple samplings conforms to the position change data of the data point in the facial action judgment model M1=1;

[0056] If the position change data of the data points corresponding to the facial action performed by the current user obtained by multiple samplings do not conform to the position change data of the data points in the facial action judgment model, M1=0;

[0057] Set up the lighting system to provide multiple color lighting changes during facial recognition;

[0058] Get real-time lighting information of all data points in the current user's facial image;

[0059] Generate facial illumination liveness judgment result M2 based on illumination color and corresponding real-time illumination information;

[0060] If the consistency evaluation value of the real-time lighting information is greater than the preset value, M2=1;

[0061] If the consistency evaluation value of the real-time lighting information is greater than the preset value, M2 = 0;

[0062] Generate a secondary liveness recognition result H2 based on the facial action liveness judgment result M1 and the facial illumination liveness judgment result M2;

[0063] Among them, when M1=1 and M2=1, H2=1; when M1≠1 or M2≠1, H2=0.

[0064] In some embodiments of the present invention, the process of generating the identification card verification result by combining the current user's identity recognition result F and the liveness recognition result H3 includes:

[0065] Based on the liveness recognition result H3, determine whether to perform identity recognition of the current user;

[0066] Repeatedly obtain the liveness recognition result H3. If H3 is always 0, the current user's identity recognition will not be performed.

[0067] If H3=1, the current user is identified and the identification result F of the current user is generated;

[0068] Generate a badge verification result based on the current user's identity recognition result F.

[0069] Compared with the prior art, the biometric-based smart sign verification and management method provided by the embodiment of the present invention has the following advantages:

[0070] By combining the liveness recognition model and the biometric information database, the accuracy and security of identity authentication are improved.

[0071] By monitoring data such as temperature, pulse, capacitance, and the judgment of facial movements and lighting in the liveness recognition model, attacks that forge biometric features are effectively prevented.

[0072] By considering the impact of ambient temperature on fingerprint temperature, a temperature range correction coefficient b is generated to adjust the temperature range, so that the model can adapt to different environmental conditions and reduce the interference of environmental factors on the recognition results.

[0073] Accurate liveness recognition is achieved through the facial liveness recognition model and the comprehensive features of facial movements and lighting, which improves the model's adaptability in various practical application scenarios.

[0074] Through illumination tests, the light reflection intensity of the test target facial data points under different colors of illumination is obtained, and the variance is calculated to generate a light reflection intensity consistency evaluation value, which helps to accurately identify facial features under different lighting conditions and reduce misidentification caused by lighting changes.

[0075] By setting facial movements in the facial recognition process, multiple sampling is performed to obtain the position change data of the data points. The facial movement judgment model is combined to generate the facial movement liveness judgment result M1, which accurately analyzes the naturalness and coherence of facial movements, helps to distinguish real facial movements from fake facial movements, and improves the accuracy of recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 This is a flow chart of a biometric-based smart sign verification management method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0077] The following embodiments of the present invention are described in further detail with reference to the accompanying drawings and examples. The following embodiments are used to illustrate the present invention but are not intended to limit the scope of the present invention.

[0078] In the description of the present invention, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention.

[0079] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the technical features being referred to. Thus, a feature specified as "first" or "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0080] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0081] Example 1:

[0082] The embodiment of the present invention provides a biometric-based smart sign verification management method, such as Figure 1 As shown, including:

[0083] Establish a liveness recognition model based on historical liveness feature data;

[0084] Combining the acquired live feature data of the current user with the live recognition model to generate the live recognition result H3 of the current user;

[0085] Enter the fingerprint feature information and facial feature information of all users to establish a biometric information database; search the biometric information database based on the acquired fingerprint feature information and facial feature information of the current user to generate the identity recognition result F of the current user;

[0086] Combine the current user's identity recognition result F and the liveness recognition result H3 to generate the identification card verification result;

[0087] The historical living body characteristic data includes: historical first living body characteristic data and historical second living body characteristic data.

[0088] Example 2:

[0089] The generation of the liveness recognition result H3 of the current user includes:

[0090] The first historical living feature data includes: historical temperature data, historical pulse data, and historical capacitance data;

[0091] Establishing a fingerprint liveness recognition model based on the first liveness feature data, and generating a first-level liveness recognition result H1 by combining the fingerprint liveness recognition model and the acquired real-time first liveness feature data;

[0092] The historical second living feature data includes: historical facial illumination data and historical facial movement data;

[0093] Establishing a facial liveness recognition model based on the second liveness feature data, and generating a secondary liveness recognition result H2 by combining the established facial liveness recognition model and the acquired real-time second liveness feature data;

[0094] Generate a liveness recognition result H3 of the current user based on the first-level liveness recognition result H1 and the second-level liveness recognition result H2;

[0095] When H1=1 and H2=1, H3=1; when H1≠1 or H2≠1, H3=0.

[0096] Example 3:

[0097] The establishment of the fingerprint liveness recognition model includes:

[0098] Historical temperature data, historical pulse data, historical capacitance data;

[0099] Generate a first living body temperature interval (a1, a2) based on the preset temperature interval of historical temperature data;

[0100] Generate a temperature interval correction coefficient b based on the current ambient temperature, and generate a second living body temperature interval (a1*(1-b), a2*(1+b)) by combining the temperature interval correction coefficient and the first living body temperature interval;

[0101] Set pulse monitoring intervals (c1, c2) based on historical pulse data;

[0102] Set the capacitance monitoring interval (d1, d2) based on historical capacitance data;

[0103] Wherein, a1 represents a first preset value of living body temperature, a2 represents a second preset value of living body temperature, c1 represents a first preset value of pulse monitoring, c2 represents a second preset value of pulse monitoring, d1 represents a first preset value of capacitance monitoring, and d2 represents a second preset value of capacitance monitoring;

[0104] A fingerprint liveness recognition model is generated by combining the second living body temperature interval (a1*(1-b), a2*(1+b)), the pulse monitoring interval (c1, c2) and the capacitance monitoring interval (d1, d2).

[0105] In this embodiment, a large amount of historical temperature data is collected, which should come from fingerprint collection processes of different individuals and different environmental conditions. Statistical analysis is performed on the historical temperature data, such as calculating statistical quantities such as mean, variance, and median.

[0106] Considering the differences in body temperature among different groups (e.g., age, gender, health status, etc.) and the impact of environmental factors (e.g., season, region, etc.) on body temperature, a reasonable temperature range is determined. For example, a normal human body temperature is generally between 36-37.5 degrees Celsius, but during the fingerprint collection process, due to factors such as skin contact with the device, there may be a certain range of fluctuations.

[0107] Based on the above analysis, the first temperature range (a1, a2) is set. The values of a1 and a2 should cover the normal temperature range of a living body while excluding abnormal temperatures. For example, a1 can be set to 35.5 degrees Celsius and a2 to 38 degrees Celsius. This range takes into account normal body temperature fluctuations as well as temperature increases that may occur due to slight exercise or environmental factors.

[0108] Generation of the temperature range correction coefficient b: Accurately measure the current ambient temperature using a high-precision temperature sensor. Develop a function based on the relationship between ambient temperature and normal human body temperature to determine the temperature range correction coefficient b.

[0109] For example, if the current ambient temperature is low, the human body's surface temperature may decrease accordingly. In this case, b is a positive value, which is used to appropriately expand the lower limit of the first living body temperature range. If the current ambient temperature is high, the body surface temperature may increase. In this case, b is also a positive value, which is used to appropriately expand the upper limit of the first living body temperature range. If the ambient temperature is within the normal comfortable range, b can be set to a value close to 0. The specific functional relationship can be determined through extensive experimental data and statistical analysis, such as b = k × (T - T0), where T is the current ambient temperature, T0 is the reference ambient temperature, and kk is a coefficient determined based on experimentation.

[0110] Calculate the second live body temperature range using the formula (a1*(1-b), a2*(1+b)). Ensure accuracy during calculation, especially with decimals. For example, if a1 = 35.5 degrees Celsius and b = 0.05, then the lower limit of the second live body temperature range is 35.5×(1-0.05) = 33.725 degrees Celsius. If a2 = 38 degrees Celsius, the upper limit is 38×(1+0.05) = 39.9 degrees Celsius.

[0111] Collect historical pulse data from different individuals in states of calm, slight exercise, and stress. Categorize and organize these data, and remove outliers (such as pulse values that are too high or too low due to measurement errors or special physiological states).

[0112] Analyze the relationship between pulse data and factors such as age, gender, and physical condition. For example, young people's pulses are generally slightly faster than those of the elderly, the pulses of men and women may also differ in some cases, and athletes' resting pulses may be lower than those of ordinary people.

[0113] Based on the above analysis, set the pulse monitoring interval (c1, c2). For example, for an average adult in a calm state, c1 can be set to 60 beats / minute and c2 can be set to 100 beats / minute. This interval takes into account the normal range of pulse fluctuations under normal physiological conditions, while also leaving some margin for possible minor movement or emotional changes.

[0114] Example 4:

[0115] The generation of the first-level living body recognition result H1 includes:

[0116] A temperature sensor, a light emitting diode, a photodetector and an AC electric field are arranged on the fingerprint sensor;

[0117] Generate user's real-time temperature data AI based on temperature sensor;

[0118] Generate user's real-time pulse data ci based on light emitting diodes and photodetectors;

[0119] Based on the AC electric field, the electric field change data when the user uses the fingerprint sensor is monitored, and the user's real-time capacitance data di is generated based on the electric field change data;

[0120] Generate real-time first living body characteristic data, including: user's real-time temperature data ai, user's real-time pulse data ci, and user's real-time capacitance data di;

[0121] Generate a first-level liveness recognition result H1 based on the generated real-time first liveness feature data;

[0122] If ai∈(a1*(1-b), a2*(1+b))) and (ci∈(c1, c2)) and (di∈(d1, d2), then H1=1;

[0123] If ai∈(1-(a1*(1-b), a2*(1+b))) or ci∈1-(c1, c2) or di∈1-(d1, d2), then H1=0.

[0124] Example 5:

[0125] The method of generating the user's real-time pulse data ci based on the light emitting diode and the photodetector includes:

[0126] The light emitted by the light emitting diode and transmitted through the finger is monitored by a photodetector;

[0127] The light intensity of the light changes periodically with the pulse beat, and the real-time pulse data ci of the user is generated.

[0128] Example 6:

[0129] The method of establishing a facial liveness recognition model based on the second liveness feature data includes:

[0130] extracting a facial feature dataset based on historical facial illumination data and historical facial motion data, wherein the facial feature dataset includes state information of a plurality of data points;

[0131] The state information of the data point includes the location information of the data point and the illumination information of the data point;

[0132] Acquire monitoring data of position information of a plurality of data points corresponding to historical facial action data to generate a facial action judgment model;

[0133] Acquire monitoring data of illumination information of a plurality of data points corresponding to historical facial illumination data to generate a facial illumination judgment model;

[0134] A facial liveness recognition model is established based on the facial action judgment model and the facial lighting judgment model.

[0135] In this embodiment, for each action (such as blinking, smiling, etc.) in the historical facial action data, position changes of multiple data points during the action are analyzed in detail.

[0136] By continuously capturing facial images or video frames, the position information of each data point at different time points is recorded to form position monitoring data. For example, for a simple blink, the coordinate position change sequence of the eye corner data point during the blink process is recorded.

[0137] Perform statistical analysis on the acquired position information monitoring data of multiple actions. Calculate the position change range, speed, acceleration and other characteristics of each data point in different actions.

[0138] Based on these features, a facial action recognition model is constructed. For example, a normal range of positional variation of data points in each action can be set. If the positional variation of data points in the monitored data falls within this range, the action is considered normal; otherwise, it is considered abnormal. The model can take the form of a probabilistic model or a rule-based model.

[0139] Based on historical facial lighting data, analyze the changes in lighting information of multiple data points under different lighting conditions (such as different colors of lighting, different intensities of lighting, etc.).

[0140] Record the monitoring data of the illumination information (such as RGB value, brightness, etc.) of each data point under different lighting conditions. For example, as the intensity of red light gradually increases, record the changes in the RGB value and brightness value of the nose tip data point.

[0141] Analyze the acquired illumination information monitoring data of multiple data points and calculate the variation pattern of illumination information under different illumination conditions, such as statistical characteristics such as variance, mean, and correlation.

[0142] Based on these statistical features, a facial illumination determination model is constructed. For example, a range of illumination variations for facial data points under certain lighting conditions can be set. If the illumination variations in the monitored data fall within this range, the illumination condition is considered normal; otherwise, it is considered abnormal. This model can take the form of a threshold-based model or a machine learning model.

[0143] Example 7:

[0144] The generation of the facial illumination judgment model includes:

[0145] Conduct lighting experiments based on the acquired historical facial lighting data;

[0146] Obtain the light reflection intensity of the test target facial data points under different colors of light;

[0147] Calculate the variance of the light reflection intensity of all facial data points of the test target under the current color illumination, and generate a light reflection intensity consistency evaluation value based on the variance of the light reflection intensity of all facial data points;

[0148] Calculate the light reflection intensity consistency evaluation value corresponding to all color lighting, and compare it with the preset value to generate the facial lighting judgment result;

[0149] A facial lighting judgment model is generated by integrating the lighting color type and the facial lighting judgment results.

[0150] In this embodiment, representative data samples are selected from a large amount of historical facial lighting data. These samples should cover facial data of people of different races, genders, and ages, as well as data under different ambient lighting conditions. For example, facial lighting data under conditions such as indoor low light, indoor strong light, and outdoor natural light should be included.

[0151] Preprocess the selected historical facial lighting data, such as normalizing the image to ensure that parameters such as image brightness and contrast are within the appropriate range to facilitate subsequent accurate analysis of lighting effects.

[0152] Based on the ambient lighting conditions in the historical facial lighting data, simulate those lighting conditions as accurately as possible in a lab environment. For example, use a lighting fixture with adjustable color and intensity, and set the lighting angle and distance to match the historical data.

[0153] Under simulated lighting conditions, facial lighting data is collected from the test target (which can be a representative facial model or a real volunteer). During the collection process, it is necessary to ensure that the parameters of the collection equipment (such as a high-precision camera) are stable and can accurately capture the lighting information of each data point on the face.

[0154] First, we need to define and select facial data points. These data points can be key facial features, such as the corners of the eyes, the tip of the nose, and the corners of the mouth, or they can be the center coordinates of facial regions defined according to specific rules. Using precise facial recognition algorithms, we can locate these data points within the image.

[0155] For each data point, a specialized light intensity measurement tool (such as a light intensity sensor integrated with the acquisition device or indirectly calculated using an image analysis algorithm) is used to obtain its light reflection intensity under different lighting colors. During the measurement process, the influence of the spectral characteristics of different lighting colors on the light reflection intensity measurement must be taken into account. For example, the reflection and absorption characteristics of skin under red light are different from those under blue light.

[0156] Record the light reflection intensity data for each facial data point under different lighting colors. Create a data table with rows representing different lighting colors and columns representing different facial data points. Each element in the table represents the corresponding light reflection intensity value. Also, record relevant environmental parameters during data collection, such as temperature and humidity, for reference during subsequent data analysis.

[0157] The light reflection intensity data of the facial data points under the current color illumination is calculated according to the variance calculation formula.

[0158] When calculating variance, it's important to pay attention to data accuracy and calculation precision. Because light reflection intensity data can be affected by a variety of factors, such as errors in the acquisition equipment and subtle differences in facial skin, multiple measurements and statistical analysis are required to ensure the reliability of the variance calculation results.

[0159] Based on the calculated variance, a light reflection intensity consistency evaluation value is generated. A mapping function can be established to map the variance value to a specific evaluation value range. For example, a small variance indicates relatively consistent light reflection intensity, and the evaluation value can be close to 1; a large variance indicates significant variation in light reflection intensity, and the evaluation value can be close to 0. This mapping function can be determined based on extensive experimental data and experience, and should be adjusted according to actual application scenarios.

[0160] Following the above method, the consistency evaluation values of the light reflection intensity under all color illuminations are calculated in sequence. For each color illumination, the steps of variance calculation and evaluation value generation are repeated to obtain a set of consistency evaluation values corresponding to different color illuminations.

[0161] These evaluation values are sorted and analyzed. For example, a bar graph can be drawn to intuitively display the distribution of evaluation values under different colors of light, so as to better understand the light reflection characteristics of facial skin under different colors of light.

[0162] Determine a reasonable preset value, which can be obtained by analyzing a large amount of normal facial lighting data. Compare the calculated light reflection intensity consistency evaluation value under all color lighting with the preset value.

[0163] If all evaluation values are greater than the preset value, it means that the facial lighting data meets the normal conditions and the facial lighting judgment result is passed, for example, it can be marked as 1; if there is at least one evaluation value less than or equal to the preset value, the facial lighting judgment result is failed and marked as 0.

[0164] The facial lighting judgment model can adopt a structured representation form, for example, it can be a matrix form, where the rows represent different lighting color types, and the columns represent the facial lighting judgment results (pass or fail) and related parameters (such as light reflection intensity variance, consistency evaluation value, etc.).

[0165] Alternatively, a rule-based representation is used to combine the illumination color type with the facial illumination judgment result and related calculation rules (such as variance calculation, evaluation value generation, preset value comparison, etc.) into a series of judgment rules.

[0166] After generating a facial lighting model, validate it using additional validation data. This validation data should be different from the historical facial lighting data used to build the model. Use this validation data to assess the model's accuracy and reliability. If errors or deficiencies are found, adjust and refine the model, such as adjusting preset values or optimizing calculation methods, until the model can accurately assess facial lighting conditions.

[0167] Example 8:

[0168] The generation of the secondary living body recognition result H2 includes:

[0169] Set the facial action in the facial recognition process and obtain the location information set E of the data points corresponding to the facial action performed by the current user, where E = {e1, e2...ei...en};

[0170] Among them, ei represents the location information of the i-th data point, and n represents the total number of data points;

[0171] The position information set E of the data points is sampled multiple times to obtain the position change data of the data points corresponding to the facial action performed by the current user;

[0172] Combining the position change data of the corresponding data points and the facial action judgment model to generate the facial action liveness judgment result M1;

[0173] If the position change data of the data point corresponding to the facial action performed by the current user obtained through multiple samplings conforms to the position change data of the data point in the facial action judgment model M1=1;

[0174] If the position change data of the data points corresponding to the facial action performed by the current user obtained by multiple samplings do not conform to the position change data of the data points in the facial action judgment model, M1=0;

[0175] Set up the lighting system to provide multiple color lighting changes during facial recognition;

[0176] Get real-time lighting information of all data points in the current user's facial image;

[0177] Generate facial illumination liveness judgment result M2 based on illumination color and corresponding real-time illumination information;

[0178] If the consistency evaluation value of the real-time lighting information is greater than the preset value, M2=1;

[0179] If the consistency evaluation value of the real-time lighting information is greater than the preset value, M2 = 0;

[0180] Generate a secondary liveness recognition result H2 based on the facial action liveness judgment result M1 and the facial illumination liveness judgment result M2;

[0181] Among them, when M1=1 and M2=1, H2=1; when M1≠1 or M2≠1, H2=0.

[0182] In this embodiment, when setting facial movements during facial recognition, consideration should be given to the ability to fully detect facial muscle movement and coordination. For example, in addition to common movements such as blinking and opening the mouth, some combined movements can also be set, such as blinking first and then smiling. For each set facial movement, a high-precision facial recognition technology (such as a facial key point detection algorithm based on deep learning) is used to obtain the position information set E = {e1, e2…ei…,en} of the data points corresponding to the current user performing the movement. These data points can be key facial feature points, such as the coordinate positions of the corners of the eyes, corners of the mouth, and the tip of the nose.

[0183] Accuracy of data point location information: To ensure the accuracy of data point location information, the image must be preprocessed during the acquisition process, such as noise removal and image enhancement. At the same time, the coordinates of the data points must be precisely quantified based on the resolution and accuracy of the facial recognition equipment.

[0184] Sampling strategy: The data point location information set E is sampled multiple times. The sampling frequency is determined by the duration and speed of the facial movement. For example, for a fast blink, a higher sampling frequency (such as 10-20 times per second) can be used, while for a relatively slow smile, a sampling frequency of 5-10 times per second can be used.

[0185] Calculation of position change data: By comparing the positions of the data points at each sampling, the position change data of the data points is calculated. For example, for the i-th data point, the difference in its coordinates between adjacent sampling moments is calculated as Δei = ei(t+1) - ei(t), where t represents the sampling moment. The position change data of all data points are combined to form a complete position change data set.

[0186] Facial Action Judgment Model Construction: The facial action judgment model is built based on a large amount of real-world facial action data. During the construction process, we collected data on the position variation of data points performing the same facial action across different populations and environments. Statistical analysis was then performed to determine the normal position variation range and pattern for each facial action.

[0187] The position change data of the data points corresponding to the facial movements performed by the current user, obtained through multiple sampling, are compared with the data in the facial movement judgment model. If these data are within the normal range determined by the model, that is, if they are consistent with the position change data of the data points in the facial movement judgment model, then M1 = 1; otherwise, M1 = 0.

[0188] Lighting system design: When configuring the lighting system to provide multiple color variations during facial recognition, it's important to select appropriate color combinations. For example, you can choose the primary colors of red, green, and blue, as well as their blends, to comprehensively test facial skin's reflective properties to different colors of light. Lighting intensity should also be appropriately set, ensuring sufficient illumination to obtain clear facial images while not being too strong to cause reflections that could affect image quality.

[0189] Acquire real-time lighting information: For all data points in the current user's facial image, obtain real-time lighting information using an image sensor (such as a camera). Lighting information can include the color component values of each data point (such as the R, G, and B values in the RGB color space) as well as the light intensity value. To improve the accuracy of lighting information, preprocessing of the acquired facial image is also required, such as color correction.

[0190] Calculate the consistency evaluation value: Calculate the consistency evaluation value of real-time lighting information. For example, a reference model can be established based on the expected reflectance characteristics of different facial regions for different colors of light. The acquired real-time lighting information is compared with the reference model, and the deviation value of each data point is calculated. The deviation values of all data points are then combined through a weighted average to obtain the consistency evaluation value.

[0191] The judgment result is generated: if the consistency evaluation value of the real-time lighting information is greater than the preset value, M2=1; if the consistency evaluation value of the real-time lighting information is less than or equal to the preset value, M2=0.

[0192] When M1=1 and M2=1, it indicates that the current user's facial movements conform to the normal pattern and the facial lighting information also conforms to expectations. At this time, H2=1, indicating that facial liveness recognition has been passed.

[0193] When M1≠1 or M2≠1, indicating an anomaly in facial motion or facial illumination, H2=0 indicates that facial liveness recognition failed. In this case, the specific issue with facial motion or facial illumination can be recorded for further analysis or to prompt the user to retry the recognition process.

[0194] Example 9:

[0195] The method of generating the identification card verification result by combining the current user's identity recognition result F and the liveness recognition result H3 includes:

[0196] Based on the liveness recognition result H3, determine whether to perform identity recognition of the current user;

[0197] Repeatedly obtain the liveness recognition result H3. If H3 is always 0, the current user's identity recognition will not be performed.

[0198] If H3=1, the current user is identified and the identification result F of the current user is generated;

[0199] Generate a badge verification result based on the current user's identity recognition result F.

[0200] In this embodiment, when the liveness recognition result H3 is 0, the badge verification result is a failure regardless of the identity recognition result F. Because the liveness detection fails, even if the identity recognition may be successful, it cannot be guaranteed to be a real user operation.

[0201] When H3 is 1 and the identity recognition result F indicates successful recognition, the identification card verification result is successful, indicating that the current user has passed the liveness detection and the identity recognition is correct.

[0202] When H3 is 1 but the identity recognition result F indicates a recognition failure, the badge verification result is a failure, indicating that although the liveness detection has passed, the identity does not match.

[0203] Finally, it should be noted that it is apparent that those skilled in the art may make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, to the extent such modifications and variations fall within the scope of the present invention and its equivalents, the present invention is intended to include such modifications and variations.

[0204] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention.

Claims

1. A biometric-based smart sign verification and management method, characterized in that: include: Establish a liveness recognition model based on historical liveness feature data; Combining the acquired live feature data of the current user with the live recognition model to generate the live recognition result H3 of the current user; Enter the fingerprint feature information and facial feature information of all users to establish a biometric information database; search the biometric information database based on the acquired fingerprint feature information and facial feature information of the current user to generate the identity recognition result F of the current user; Combine the current user's identity recognition result F and the liveness recognition result H3 to generate the identification card verification result; The historical living body characteristic data includes: historical first living body characteristic data and historical second living body characteristic data.

2. The biometric-based smart sign verification management method according to claim 1, characterized in that: The generation of the liveness recognition result H3 of the current user includes: The first historical living feature data includes: historical temperature data, historical pulse data, and historical capacitance data; Establishing a fingerprint liveness recognition model based on the first liveness feature data, and generating a first-level liveness recognition result H1 by combining the fingerprint liveness recognition model and the acquired real-time first liveness feature data; The historical second living feature data includes: historical facial illumination data and historical facial movement data; Establishing a facial liveness recognition model based on the second liveness feature data, and generating a secondary liveness recognition result H2 by combining the established facial liveness recognition model and the acquired real-time second liveness feature data; Generate a liveness recognition result H3 of the current user based on the first-level liveness recognition result H1 and the second-level liveness recognition result H2; When H1=1 and H2=1, H3=1; when H1≠1 or H2≠1, H3=0.

3. The biometric-based smart sign verification management method according to claim 2, characterized in that: The establishment of the fingerprint liveness recognition model includes: Historical temperature data, historical pulse data, historical capacitance data; Generate a first living body temperature interval (a1, a2) based on the preset temperature interval of historical temperature data; Generate a temperature interval correction coefficient b based on the current ambient temperature, and generate a second living body temperature interval (a1*(1-b), a2*(1+b)) by combining the temperature interval correction coefficient and the first living body temperature interval; Set pulse monitoring intervals (c1, c2) based on historical pulse data; Set the capacitance monitoring interval (d1, d2) based on historical capacitance data; Wherein, a1 represents a first preset value of living body temperature, a2 represents a second preset value of living body temperature, c1 represents a first preset value of pulse monitoring, c2 represents a second preset value of pulse monitoring, d1 represents a first preset value of capacitance monitoring, and d2 represents a second preset value of capacitance monitoring; A fingerprint liveness recognition model is generated by combining the second living body temperature interval (a1*(1-b), a2*(1+b)), the pulse monitoring interval (c1, c2) and the capacitance monitoring interval (d1, d2).

4. The biometric-based smart sign verification management method according to claim 3, characterized in that: The generation of the first-level living body recognition result H1 includes: A temperature sensor, a light emitting diode, a photodetector and an AC electric field are arranged on the fingerprint sensor; Generate user's real-time temperature data AI based on temperature sensor; Generate user's real-time pulse data ci based on light emitting diodes and photodetectors; Based on the AC electric field, the electric field change data when the user uses the fingerprint sensor is monitored, and the user's real-time capacitance data di is generated based on the electric field change data; Generate real-time first living body characteristic data, including: user's real-time temperature data ai, user's real-time pulse data ci, and user's real-time capacitance data di; Generate a first-level liveness recognition result H1 based on the generated real-time first liveness feature data; If ai∈(a1*(1-b), a2*(1+b))) and (ci∈(c1, c2)) and (di∈(d1, d2), then H1=1; If ai∈(1-(a1*(1-b), a2*(1+b))) or ci∈1-(c1, c2) or di∈1-(d1, d2), then H1=0.

5. The biometric-based smart sign verification management method according to claim 4, characterized in that: The method of generating the user's real-time pulse data ci based on the light emitting diode and the photodetector includes: The light emitted by the light emitting diode and transmitted through the finger is monitored by a photodetector; The light intensity of the light changes periodically with the pulse beat, and the real-time pulse data ci of the user is generated.

6. The biometric-based smart sign verification management method according to claim 5, characterized in that: The method of establishing a facial liveness recognition model based on the second liveness feature data includes: extracting a facial feature dataset based on historical facial illumination data and historical facial motion data, wherein the facial feature dataset includes state information of a plurality of data points; The state information of the data point includes the location information of the data point and the illumination information of the data point; Acquire monitoring data of position information of a plurality of data points corresponding to historical facial action data to generate a facial action judgment model; Acquire monitoring data of illumination information of a plurality of data points corresponding to historical facial illumination data to generate a facial illumination judgment model; A facial liveness recognition model is established based on the facial action judgment model and the facial lighting judgment model.

7. The biometric-based smart sign verification management method according to claim 6, characterized in that: The generation of the facial illumination judgment model includes: Conduct lighting experiments based on the acquired historical facial lighting data; Obtain the light reflection intensity of the test target facial data points under different colors of light; Calculate the variance of the light reflection intensity of all facial data points of the test target under the current color illumination, and generate a light reflection intensity consistency evaluation value based on the variance of the light reflection intensity of all facial data points; Calculate the light reflection intensity consistency evaluation value corresponding to all color lighting, and compare it with the preset value to generate the facial lighting judgment result; A facial lighting judgment model is generated by integrating the lighting color type and the facial lighting judgment results.

8. The biometric-based smart sign verification management method according to claim 7, characterized in that: The generation of the secondary living body recognition result H2 includes: Set the facial action in the facial recognition process and obtain the location information set E of the data points corresponding to the facial action performed by the current user, where E = {e1, e2...ei...en}; Among them, ei represents the location information of the i-th data point, and n represents the total number of data points; The position information set E of the data points is sampled multiple times to obtain the position change data of the data points corresponding to the facial action performed by the current user; Combining the position change data of the corresponding data points to generate a facial action liveness judgment result M1 based on the facial action judgment model; If the position change data of the data point corresponding to the facial action performed by the current user obtained through multiple samplings conforms to the position change data of the data point in the facial action judgment model M1=1; If the position change data of the data points corresponding to the facial action performed by the current user obtained by multiple samplings do not conform to the position change data of the data points in the facial action judgment model, M1=0; Set up the lighting system to provide multiple color lighting changes during facial recognition; Get real-time lighting information of all data points in the current user's facial image; Generate facial illumination liveness judgment result M2 based on illumination color and corresponding real-time illumination information; If the consistency evaluation value of the real-time lighting information is greater than the preset value, M2=1; If the consistency evaluation value of the real-time lighting information is greater than the preset value, M2 = 0; Generate a secondary liveness recognition result H2 based on the facial action liveness judgment result M1 and the facial illumination liveness judgment result M2; Among them, when M1=1 and M2=1, H2=1; when M1≠1 or M2≠1, H2=0.

9. The biometric-based smart sign verification management method according to claim 8, characterized in that: The method of generating the identification card verification result by combining the current user's identity recognition result F and the liveness recognition result H3 includes: Based on the liveness recognition result H3, determine whether to perform identity recognition of the current user; Repeatedly obtain the liveness recognition result H3. If H3 is always 0, the current user's identity recognition will not be performed. If H3=1, the current user is identified and the identification result F of the current user is generated; Generate a badge verification result based on the current user's identity recognition result F.

Images