An artificial intelligence-based multi-modal biometric recognition method

Abstract

The application discloses a kind of multi-modal biological recognition methods based on artificial intelligence, it is related to biological recognition field, the problem that existing biological recognition method exists is solved to the problem of poor recognition effect, including steps S1: to each outdoor biological recognition area is carried out inverse light intensity analysis and inverse light angle analysis, obtain region inverse light analysis data according to analysis result, step S2: the outdoor biological recognition area is divided into first type biological recognition area and second type biological recognition area, obtain recognition area type division data, and biological recognition is carried out to first type biological recognition area, step S3: biological palmprint image analysis is carried out to the biological to be identified in second type biological recognition area, and biological recognition is carried out according to analysis result, the accuracy of the present application can improve multi-modal biological recognition method.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of biometric identification, and relates to an artificial intelligence technology, in particular to a multi-modal biometric identification method based on artificial intelligence. BACKGROUND

[0002] The existing multi-modal biometric identification method has the following defects when performing biometric identification:

[0003] 1. The existing multi-modal biometric identification method cannot perform back light intensity analysis and back light angle analysis on an outdoor biometric identification area to obtain a back light identification influence coefficient for selection of an identification mode, resulting in a lag in the identification process.

[0004] 2. The existing multi-modal biometric identification method can only perform real-time comparison between a pre-recorded palm print image and a palm print image to be identified when performing biometric palm print identification, and cannot train an image model in combination with historical data, thereby resulting in low biometric identification efficiency.

[0005] Therefore, the application provides a multi-modal biometric identification method based on artificial intelligence. SUMMARY

[0006] In view of the deficiencies of the prior art, the application aims to provide a multi-modal biometric identification method based on artificial intelligence, and aims to improve the identification efficiency and identification accuracy of the multi-modal biometric identification method.

[0007] In order to achieve the above-mentioned purpose, the application adopts the following technical solution: a multi-modal biometric identification method based on artificial intelligence, comprising the following steps:

[0008] Step S1: Obtain a plurality of outdoor biometric identification areas, perform back light intensity analysis and back light angle analysis on each outdoor biometric identification area respectively, obtain a back light identification influence coefficient corresponding to each outdoor biometric identification area according to the analysis results, and obtain area back light analysis data.

[0009] Step S2: Perform identification mode analysis on the outdoor biometric identification area according to the area back light analysis data, divide the outdoor biometric identification area into a first type biometric identification area and a second type biometric identification area according to the analysis results, obtain identification area type division data, and perform biometric identification on the first type biometric identification area using a first biometric identification device.

[0010] Step S3: Obtain the second type biometric identification area according to the identification area type division data, perform biometric palm print image analysis on a biometric object to be identified in the second type biometric identification area using a second biometric identification device, and perform biometric identification according to the analysis results.

[0011] Further, the step S1 further includes the following specific steps.

[0012] Step S11: Acquire the outdoor area needing to be biologically identified, obtain a plurality of outdoor biological identification areas, select a sample biological identification area in the acquired outdoor biological identification area, and set two biological identification devices in the sample biological identification area respectively, and name them as the first biological identification device and the second biological identification device respectively;

[0013] Step S12: Perform biological facial light-reversal degree analysis on the sample biological identification area, and obtain a biological light-reversal identification influence coefficient according to the analysis result.

[0014] Step S13: Obtain the light-reversal identification influence coefficient corresponding to each outdoor biological identification area respectively, and obtain regional light-reversal analysis data.

[0015] Further, the step S12 further includes the following specific steps.

[0016] Step S121: If there is a biological body to be identified in the sample biological identification area, create an identification area space coordinate system according to the relative position of the biological body to be identified and the first biological identification device;

[0017] Step S122: Perform facial light-reversal angle analysis on the biological body to be identified according to the identification area space coordinate system, and obtain a facial light-reversal included angle corresponding to the biological body to be identified according to the analysis result.

[0018] Step S123: Numerically acquire the light intensity corresponding to the sample biological identification area at the current time, and obtain a regional real-time light intensity value.

[0019] Step S124: Calculate the biological light-reversal identification influence coefficient corresponding to the sample biological identification area by using the facial light-reversal included angle corresponding to the biological body to be identified and the regional real-time light intensity value.

[0020] The biological light-reversal identification influence coefficient corresponding to the sample biological identification area is calculated, and the specific formula is as follows:

[0021]

[0022] Wherein, Nxk is the biological light-reversal identification influence coefficient corresponding to the sample biological identification area, Mjj is the facial light-reversal included angle corresponding to the biological body to be identified, Lss is the regional real-time light intensity value, and Lck is the reference light intensity value.

[0023] Further, the step S121 further includes the following specific steps.

[0024] Within the sample biometric area, the geometric center point of the face corresponding to the organism to be identified is acquired by the image acquisition device to obtain the first position feature point. A horizontal straight line pointing due east is drawn through the first position feature point to obtain the first position feature line. A vertical straight line pointing to the sky is drawn through the first position feature point to obtain the second position feature line. A horizontal straight line pointing due north is drawn through the first position feature point to obtain the third position feature line.

[0025] The first position feature point is marked as the origin of the coordinate system, the first position feature line is marked as the x-axis, the second position feature line is marked as the y-axis, and the third position feature line is marked as the z-axis, thus obtaining the spatial coordinate system of the recognition area.

[0026] Furthermore, step S122 also includes the following specific steps:

[0027] Step S1221: Perform position analysis on the first biometric device according to the spatial coordinate system of the identification area, and obtain the device's line-of-sight vector based on the analysis results;

[0028] Step S1222: Analyze the position of sunlight based on the spatial coordinate system of the identified area, and obtain the sunlight vector based on the analysis results.

[0029] Step S1223: Vector the sunlight and the device's line-of-sight vector The backlight angle of the face corresponding to the organism to be identified is obtained by calculation;

[0030] The backlight angle corresponding to the face of the organism to be identified is calculated using the following formula:

[0031]

[0032] Where Mjj is the backlight angle of the face of the organism to be identified. The vector of sunlight. This is the line-of-sight vector of the device.

[0033] Furthermore, step S1221 further includes the following specific steps:

[0034] The horizontal distance between the center point of the first biometric device and the geometric center point of the organism's face is obtained by identifying the spatial coordinate system of the region, thus obtaining the first horizontal distance value.

[0035] The vertical distance between the center point of the first biometric device and the geometric center point of the organism's face is obtained by identifying the spatial coordinate system of the region, thus obtaining the first vertical distance value.

[0036] The angle between the horizontal ray passing through the center point of the first biometric device and the vertical plane directly in front of the face of the organism is obtained by identifying the spatial coordinate system of the region; thus, the horizontal offset angle of the first device is obtained.

[0037] The pitch angle of the first biometric device is obtained by identifying the angle between the horizontal ray passing through the center point of the first biometric device and the horizontal plane where the geometric center point of the organism's face is located by the spatial coordinate system of the identification area.

[0038] The vector pointing from the center point of the first biometric device to the geometric center point of the organism's face is labeled as the device's line-of-sight vector.

[0039] Acquisition device line-of-sight vector The projection length in the XZ plane is: dxz = d·cosε;

[0040] Acquisition device line-of-sight vector The corresponding x-axis component: Cx = dxz·sinδ;

[0041] Acquisition device line-of-sight vector The corresponding z-axis component: Cz = dxz·cosδ;

[0042] Acquisition device line-of-sight vector The corresponding y-axis component: Cy = d·cosε;

[0043] Where d is the first horizontal distance value, h is the first vertical distance value, ε is the first device pitch angle, and δ is the first device horizontal offset angle;

[0044] The device's line-of-sight vector is obtained by calculating the first horizontal distance value d and the first vertical distance value h. Length of the module

[0045] Line of sight through the device The corresponding x-axis component Cx, z-axis component Cz, y-axis component Cy, and device line-of-sight vector Length of the module Line of sight vector of the device The calculation is performed using the following formula:

[0046]

[0047] Furthermore, step S1222 further includes the following specific steps:

[0048] Obtain the solar altitude angle α corresponding to the current moment, obtain the angle of the sun relative to due east at the current moment, and obtain the sun's eastward azimuth angle β;

[0049] Obtain the sun ray vector Projection length in the coordinate XZ plane: cos a;

[0050] Obtain the sun ray vector The corresponding x-axis component: Sx = cos a sin b;

[0051] Obtain the sun ray vector The corresponding z-axis component: Sz = cos a cos b;

[0052] Obtain the sun ray vector The corresponding y-axis component: Sy = sin a;

[0053] Then, through the sun ray vector The projection length in the coordinate XZ plane, the x-axis component, the z-axis component, and the y-axis component of the sun ray vector can be calculated The specific formula is as follows:

[0054]

[0055] Further, the step S2 further includes the following specific steps:

[0056] Step S21: Obtain the area backlight analysis data, and obtain the backlight recognition influence coefficient corresponding to each outdoor biological recognition area according to the area backlight analysis data;

[0057] Step S22: Obtain the backlight recognition influence coefficient reference interval, if the backlight recognition influence coefficient is in the backlight recognition influence coefficient reference interval, the corresponding outdoor biological recognition area is divided into a first type of biological recognition area, if the backlight recognition influence coefficient is not in the backlight recognition influence coefficient reference interval, the corresponding outdoor biological recognition area is divided into a second type of biological recognition area, and recognition area type division data is obtained;

[0058] Step S23: The biological face of the first type of biological recognition area is collected by the first biological recognition device, and the collection result is compared with the existing biological face image information library for image consistency, if the comparison passes, the biological recognition is successful, if the comparison fails, the biological recognition fails.

[0059] Further, the step S3 further includes the following steps:

[0060] Step S31: Obtain the recognition area type division data, and obtain the second type of biological recognition area according to the recognition area type division data;

[0061] Step S32: Obtain the biological palmprint image of the to-be-recognized biological in the second type of biological recognition area, and obtain the to-be-recognized palmprint image;

[0062] Step S33: Obtain the access palmprint images stored in the second biometric device, obtain a plurality of stored palmprint images, create a palmprint recognition model according to the plurality of stored palmprint images, and obtain a feature palmprint recognition model;

[0063] Step S34: Identify the to-be-identified palmprint image using the feature palmprint recognition model. If the output result of the feature palmprint recognition model is that the image comparison passes, the biometric recognition is completed. If the output result of the feature palmprint recognition model is that the image comparison fails, the biometric recognition fails.

[0064] Further, the step S33 further includes the following steps:

[0065] A model training image acquisition period is created by taking the time point corresponding to the current time as the period end time point. The palmprint images obtained by the second biometric device in the model training image acquisition period are obtained, and a plurality of palmprint training images are obtained.

[0066] Any sample palmprint training image is selected from the plurality of palmprint training images, and the sample palmprint training image is matched with the stored palmprint image for consistency. The sample palmprint training image is identified according to the matching result.

[0067] Specifically as follows:

[0068] Any sample stored palmprint image is selected from the plurality of stored palmprint images. The palm public region in the sample palmprint training image and the sample stored palmprint image is obtained by an image recognition algorithm. The palm public region in the sample palmprint training image is marked as a first palm public region, and the palm public region in the sample stored palmprint image is marked as a second palm public region.

[0069] The first palm public region and the second palm public region are respectively subjected to image grayscale processing to obtain a first grayscale image and a second grayscale image.

[0070] The palmprint coverage region in the second grayscale image is marked to obtain a second image palmprint coverage region, and the second image palmprint coverage region is disassembled into a plurality of pixel points. The pixel depth corresponding to each pixel point is obtained to obtain a plurality of pixel depth values. The maximum pixel depth value is marked as a first palmprint pixel depth reference value, and the minimum pixel depth value is marked as a second palmprint pixel depth reference value. The first palmprint pixel depth reference value and the second palmprint pixel depth reference value are marked as a palmprint pixel point reference interval.

[0071] Split the first gray image into a plurality of pixel points, and obtain the pixel depth corresponding to each pixel point to obtain a plurality of pixel depth values, and mark the pixel points with pixel depth values in the palm print pixel point reference interval as palm print region pixel points;

[0072] In the first gray image, mark the image region occupied by the palm print region pixel points as the first image palm print coverage region;

[0073] Use the first gray image to cover the second gray image, name the overlapping region of the first image palm print coverage region and the second image palm print coverage region as the image palm print overlapping region, and obtain the area of the image palm print overlapping region to obtain the image palm print overlapping region area value;

[0074] Obtain the area value of the second image palm print coverage region, obtain the palm print region area value corresponding to the sample storage palm print image, obtain the storage palm print region area value, calculate the ratio of the image palm print overlapping region area value to the storage palm print region area value, and obtain the palm print region similarity between the sample palm print training image and the sample storage palm print image;

[0075] Obtain the palm print region similarity between the sample palm print training image and each storage palm print image, and mark the palm print region similarity with the maximum value as the comprehensive comparison similarity corresponding to the sample palm print training image;

[0076] Obtain the similarity qualified preset interval, if the comprehensive comparison similarity is in the similarity qualified preset interval, mark the sample palm print training image as comparison passed, if the comprehensive comparison similarity is not in the similarity qualified preset interval, mark the sample palm print training image as comparison failed;

[0077] Identify each palm print training image respectively to obtain palm print training image marking data;

[0078] Create an image recognition model through the existing artificial intelligence platform, train the image recognition model using the palm print training image marking data, and obtain a feature palm print recognition model.

[0079] As described above, due to the adoption of the above technical solutions, the beneficial effects of the present application are:

[0080] 1. The present application can ensure the accuracy of the identification process by analyzing the inverse light intensity and the inverse light angle of the outdoor biological recognition region to obtain the inverse light recognition influence coefficient, and selecting the identification mode according to the inverse light recognition influence coefficient;

[0081] 2、The application can effectively improve the biological recognition efficiency by combining historical images to create a feature palmprint recognition model and performing biological palmprint recognition through the feature palmprint recognition model, compared with the traditional recognition method of comparing the pre-recorded palmprint image with the to-be-recognized palmprint image in real time. BRIEF DESCRIPTION OF DRAWINGS

[0082] In order to facilitate the understanding of those skilled in the art, the present application will be further described below in conjunction with the drawings.

[0083] Figure 1 It is the overall system block diagram of the present application;

[0084] Figure 2 It is the recognition area space coordinate system schematic diagram in the present application;

[0085] Figure 3 It is the face backlight angle schematic diagram in the present application;

[0086] Figure 4 It is the palmprint coverage area schematic diagram in the present application. DETAILED DESCRIPTION

[0087] The technical solutions of the present application will be described below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0088] Embodiment one

[0089] Please refer to Figure 1 The present application provides a technical solution: a multi-modal biological recognition method based on artificial intelligence, comprising the following specific steps:

[0090] Step S1: obtaining a plurality of outdoor biological recognition areas, respectively analyzing the inverse light intensity and the inverse light angle of each outdoor biological recognition area, obtaining the inverse light recognition influence coefficient corresponding to each outdoor biological recognition area according to the analysis results, and obtaining the area inverse light analysis data;

[0091] The step S1 further comprises the following specific steps:

[0092] The outdoor area needing biological recognition is obtained to obtain a plurality of outdoor biological recognition areas, and a sample biological recognition area is selected in the obtained outdoor biological recognition area, and two biological recognition devices are respectively arranged in the sample biological recognition area, and are respectively named as the first biological recognition device and the second biological recognition device;

[0093] It should be noted that:

[0094] In the present application, the first biological recognition device referred to here is a face recognition device, and the second biological recognition device is a palmprint recognition device;

[0095] Performing biological face inverse light degree analysis on the sample biological recognition area, and obtaining a biological inverse light recognition influence coefficient according to the analysis result;

[0096] Specifically as follows:

[0097] If there is a biological body to be recognized in the sample biological recognition area, then a recognition area space coordinate system is created according to the relative position between the biological body to be recognized and the first biological recognition device;

[0098] Specifically as follows:

[0099] Please refer to Figure 2 In the sample biological recognition area, the face geometric center point corresponding to the biological body to be recognized is obtained by the image acquisition device, to obtain a first position feature point, a straight line is drawn through the first position feature point and pointing horizontally to the east to obtain a first position feature straight line, a straight line is drawn through the first position feature point and vertically to the sky to obtain a second position feature straight line, and a straight line is drawn through the first position feature point and horizontally to the north to obtain a third position feature straight line;

[0100] It should be noted here that:

[0101] In the present application, the sample biological recognition area referred to here is an outdoor area.

[0102] The first position feature point is marked as the coordinate origin, the first position feature straight line is marked as the coordinate x-axis, the second position feature straight line is marked as the coordinate y-axis, and the third position feature straight line is marked as the coordinate z-axis to obtain the recognition area space coordinate system;

[0103] Performing face inverse light angle analysis on the biological body to be recognized according to the recognition area space coordinate system, and obtaining a face inverse light included angle corresponding to the biological body to be recognized according to the analysis result;

[0104] Specifically as follows:

[0105] Performing position analysis on the first biological recognition device according to the recognition area space coordinate system, and obtaining a device line-of-sight vector according to the analysis result;

[0106] Specifically as follows:

[0107] Please refer to Figure 3 The horizontal distance value between the center point of the first biological recognition device and the face geometric center point of the biological body is obtained through the recognition area space coordinate system to obtain a first horizontal distance value;

[0108] The vertical distance between the center point of the first biometric device and the geometric center point of the organism's face is obtained by identifying the spatial coordinate system of the region, thus obtaining the first vertical distance value.

[0109] The angle between the horizontal ray passing through the center point of the first biometric device and the vertical plane directly in front of the face of the organism is obtained by identifying the spatial coordinate system of the region; thus, the horizontal offset angle of the first device is obtained.

[0110] It should be noted here that: if the center point of the first biometric device is on the right side of the organism's face, the horizontal offset angle of the first device is set to a positive value; if the center point of the first biometric device is on the left side of the organism's face, the horizontal offset angle of the first device is set to a negative value.

[0111] The pitch angle of the first biometric device is obtained by identifying the angle between the horizontal ray passing through the center point of the first biometric device and the horizontal plane where the geometric center point of the organism's face is located, using the spatial coordinate system of the identification area.

[0112] The vector pointing from the center point of the first biometric device to the geometric center point of the organism's face is labeled as the device's line-of-sight vector.

[0113] Acquisition device line-of-sight vector The projection length in the XZ plane is: dxz = d·cosε;

[0114] Acquisition device line-of-sight vector The corresponding x-axis component: Cx = dxz·sinδ;

[0115] Acquisition device line-of-sight vector The corresponding z-axis component: Cz = dxz·cosδ;

[0116] Acquisition device line-of-sight vector The corresponding y-axis component: Cy = d·cosε;

[0117] Where d is the first horizontal distance value, h is the first vertical distance value, ε is the first device pitch angle, and δ is the first device horizontal offset angle;

[0118] The device's line-of-sight vector is obtained by calculating the first horizontal distance value d and the first vertical distance value h. Length of the module

[0119] Line of sight through the device The corresponding x-axis component Cx, z-axis component Cz, y-axis component Cy, and device line-of-sight vector Length of the module Line of sight vector of the device The calculation is performed using the following formula:

[0120]

[0121] It should be noted here that:

[0122] In the specific implementation, if d = 1.2m, h = 0.15m, δ = 10°,

[0123] Then ε = arctan (0.15 / 1.2≈7.13°

[0124] Device line-of-sight vector

[0125] According to the position analysis of the sunlight ray according to the identification area spatial coordinate system, the sunlight ray vector is obtained according to the analysis result

[0126] Specifically as follows:

[0127] Obtain the solar elevation angle α corresponding to the current moment, obtain the angle of the sun relative to the positive east direction at the current moment, and obtain the solar east azimuth angle β;

[0128] Obtain the sunlight ray vector The projection length in the coordinate XZ plane: cosα;

[0129] Obtain the sunlight ray vector The x-axis component corresponding to: Sx = cosα·sinβ;

[0130] Obtain the sunlight ray vector The z-axis component corresponding to: Sz = cosα·cosβ;

[0131] Obtain the sunlight ray vector The y-axis component corresponding to: Sy = sinα;

[0132] Then the sunlight ray vector The projection length in the coordinate XZ plane, the x-axis component, the z-axis component, and the y-axis component of the sunlight ray vector The specific formula is as follows:

[0133]

[0134] It should be noted here that:

[0135] In the specific implementation, if α = 73.2°, β = 190.5°, the sunlight ray vector

[0136] The sunlight ray vector And the device line-of-sight vector The face inverse light included angle corresponding to the to-be-identified biological is obtained by calculation;

[0137] The face inverse light included angle corresponding to the to-be-identified biological is calculated, and the specific formula is as follows:

[0138]

[0139] Wherein, Mjj is the face inverse light included angle corresponding to the to-be-identified biological, is the sun light vector, is the device line-of-sight vector;

[0140] The light intensity corresponding to the sample biological identification region at the current moment is obtained, and the region real-time light intensity value is obtained.

[0141] The biological inverse light identification influence coefficient corresponding to the sample biological identification region is obtained by calculation of the face inverse light included angle corresponding to the to-be-identified biological and the region real-time light intensity value.

[0142] The inverse light identification influence coefficient corresponding to the sample biological identification region is calculated, and the specific formula is as follows:

[0143]

[0144] Wherein, Nxk is the biological inverse light identification influence coefficient corresponding to the sample biological identification region, Mjj is the face inverse light included angle corresponding to the to-be-identified biological, Lss is the region real-time light intensity value, and Lck is the reference light intensity value.

[0145] It should be noted here that:

[0146] In the present application, the reference light intensity value involved here is a constant, and in the specific implementation, the reference light intensity value is set to 50000 lux.

[0147] In the specific implementation, when the face inverse light included angle corresponding to the to-be-identified biological is 120°, and the region real-time light intensity value corresponding to the sample biological identification region is 120000 lux, the biological inverse light identification influence coefficient corresponding to the sample biological identification region can be calculated as 0.7059.

[0148] It should be noted here that:

[0149] The existing identification of the to-be-identified biological body by the first biological identification device often needs the to-be-identified biological body to be close to the identification device, and direct face recognition is performed on the to-be-identified biological body until the identification fails, and even multiple identification failures will use the second biological identification device to identify the to-be-identified biological body, the present application can further reduce the identification time and improve the identification efficiency by using the biological backlight recognition influence coefficient to make a preliminary judgment on whether the first biological identification device can be identified, and according to the judgment result, it is decided whether to use the first biological identification device or the second biological identification device to identify the to-be-identified biological body.

[0150] The acquisition process of the backlight recognition influence coefficient corresponding to the sample biological identification region is repeated to acquire the backlight recognition influence coefficient corresponding to each outdoor biological identification region, and the region backlight analysis data is obtained.

[0151] Step S2: According to the region backlight analysis data, the outdoor biological identification region is analyzed, and according to the analysis result, the outdoor biological identification region is divided into a first type biological identification region and a second type biological identification region, and the identification region type division data is obtained, and the first biological identification device is used to identify the first type biological identification region.

[0152] In the step S2, the following specific steps are further included:

[0153] The region backlight analysis data is acquired, and the backlight recognition influence coefficient corresponding to each outdoor biological identification region is acquired according to the region backlight analysis data.

[0154] The backlight recognition influence coefficient reference interval is acquired, if the backlight recognition influence coefficient is in the backlight recognition influence coefficient reference interval, the corresponding outdoor biological identification region is divided into the first type biological identification region, if the backlight recognition influence coefficient is not in the backlight recognition influence coefficient reference interval, the corresponding outdoor biological identification region is divided into the second type biological identification region, and the identification region type division data is obtained.

[0155] It should be noted here that:

[0156] The backlight recognition influence coefficient reference interval is acquired, and the specific steps are as follows:

[0157] Obtaining a plurality of historical identification records of biological identification completed by the first biological identification device, obtaining a plurality of inverse light identification influence coefficients corresponding to each historical identification record respectively, comparing the numerical values of the plurality of inverse light identification influence coefficients, marking the inverse light identification influence coefficient with the largest numerical value as a first inverse light identification influence coefficient reference value, marking the inverse light identification influence coefficient with the smallest numerical value as a second inverse light identification influence coefficient reference value, and marking the numerical interval composed of the first inverse light identification influence coefficient reference value and the second inverse light identification influence coefficient as an inverse light identification influence coefficient reference interval;

[0158] Collecting the biological face of the first type biological identification area by the first biological identification device, and performing image consistency comparison between the collection result and the existing biological face image information library. If the comparison passes, the biological identification is successful, and if the comparison fails, the biological identification fails.

[0159] It should be noted here that:

[0160] The first biological identification device involved here can be an outdoor face recognition access control. The biological face identification performed by the first biological identification device can be human face recognition. If the identification is successful, the access control passes, and if the identification is not successful, the access control fails.

[0161] Step S3: According to the identification area type division data, a second type biological identification area is obtained, and a second biological identification device is used to analyze the biological palmprint image of the to-be-identified biological in the second type biological identification area, and biological identification is performed according to the analysis result.

[0162] The step S3 further includes the following steps:

[0163] Obtaining identification area type division data, obtaining a second type biological identification area according to the identification area type division data, obtaining a plurality of second type biological identification areas, and selecting a characteristic biological identification area from the plurality of second type biological identification areas;

[0164] Performing biological palmprint analysis on the to-be-identified biological in the characteristic biological identification area, and obtaining a biological palmprint collection image according to the analysis result;

[0165] Specifically as follows:

[0166] Obtaining a biological palmprint image corresponding to the to-be-identified biological, and obtaining a to-be-identified palmprint image;

[0167] Obtaining the access palmprint images stored in the second biological identification device, obtaining a plurality of stored palmprint images, creating a palmprint identification model according to the plurality of stored palmprint images, and obtaining a characteristic palmprint identification model;

[0168] Specifically as follows:

[0169] A model training image acquisition cycle is created with the time point corresponding to the current time as the cycle end time point, and palmprint images acquired by the second biological recognition device in the model training image acquisition cycle are acquired to obtain a plurality of palmprint training images;

[0170] It should be noted here that:

[0171] In the present application, the cycle length corresponding to the model training image acquisition cycle is one month. If the time interval between the second biological recognition device and the current time point is less than or equal to one month, the time point at which the first biological recognition device is put into use is marked as the cycle start time point. If the time interval between the second biological recognition device and the current time point is greater than one month, the time point corresponding to one month before the cycle end time point is marked as the cycle start time point.

[0172] An arbitrary sample palmprint training image is selected from the plurality of acquired palmprint training images, and the sample palmprint training image is matched with the stored palmprint image for consistency. The sample palmprint training image is identified and marked according to the matching result.

[0173] Specifically as follows:

[0174] An arbitrary sample stored palmprint image is selected from the plurality of stored palmprint images, and the palm common area in the sample palmprint training image and the sample stored palmprint image is acquired by an image recognition algorithm. The palm common area in the sample palmprint training image is marked as a first palm common area, and the palm common area in the sample stored palmprint image is marked as a second palm common area.

[0175] The first palm common area and the second palm common area are respectively subjected to image grayscale processing to obtain a first grayscale image and a second grayscale image.

[0176] It should be noted here that:

[0177] The first grayscale image and the second grayscale image correspond to the same image parameters, which include but are not limited to image size, image brightness, and image resolution.

[0178] Mark the palm print coverage area in the second grayscale image to obtain a second image palm print coverage area, and disassemble the second image palm print coverage area into a plurality of pixel points, respectively acquire the pixel depth corresponding to each pixel point to obtain a plurality of pixel depth values, mark the maximum pixel depth value as a first palm print pixel depth reference value, mark the minimum pixel depth value as a second palm print pixel depth reference value, and mark the first palm print pixel depth reference value and the second palm print pixel depth reference value as a palm print pixel point reference interval;

[0179] It should be noted here that:

[0180] Please refer to Figure 4 In this application, the palm print coverage area in the second grayscale image referred to here is specifically manually marked by the sample storage palm print image during the input process.

[0181] Disassemble the first grayscale image into a plurality of pixel points, respectively acquire the pixel depth corresponding to each pixel point to obtain a plurality of pixel depth values, and mark the pixel points with pixel depth values in the palm print pixel point reference interval as palm print region pixel points;

[0182] In the first grayscale image, mark the image area occupied by the palm print region pixel points as a first image palm print coverage area;

[0183] Use the first grayscale image to cover the second grayscale image, name the overlapping area of the first image palm print coverage area and the second image palm print coverage area as an image palm print overlapping area, and acquire the area of the image palm print overlapping area to obtain an image palm print overlapping area value;

[0184] Acquire the area value of the second image palm print coverage area to obtain the palm print area value corresponding to the sample storage palm print image, obtain the storage palm print area value, calculate the ratio of the image palm print overlapping area value to the storage palm print area value, and obtain the palm print area similarity between the sample palm print training image and the sample storage palm print image;

[0185] Repeat the process of acquiring the palm print area similarity between the sample palm print training image and the sample storage palm print image, respectively acquire the palm print area similarity between the sample palm print training image and each storage palm print image, and mark the palm print area similarity with the largest value as the comprehensive comparison similarity corresponding to the sample palm print training image;

[0186] Acquire a similarity qualified preset interval, if the comprehensive comparison similarity is in the similarity qualified preset interval, mark the sample palm print training image as comparison passed, if the comprehensive comparison similarity is not in the similarity qualified preset interval, mark the sample palm print training image as comparison failed;

[0187] It should be noted here that:

[0188] In the present application, the similarity qualified preset interval is obtained as follows:

[0189] A plurality of comparison passed historical palmprint images are obtained, the palmprint region similarity corresponding to each historical palmprint image is obtained respectively, and the plurality of palmprint region similarities are compared in value size, the palmprint region similarity with the largest value is marked as the upper limit of the similarity qualified preset interval, and the palmprint region similarity with the smallest value is marked as the lower limit of the similarity qualified preset interval, to obtain the similarity qualified preset interval;

[0190] The sample palmprint training image is repeatedly identified and marked, each palmprint training image is identified and marked respectively, and palmprint training image marking data is obtained;

[0191] It should be noted here that:

[0192] The process of dividing the historical palmprint image into "comparison not passed" and "comparison passed" is that the model automatically extracts pixel points from the first grayscale image, covers the image, and analyzes the area value. This process realizes the comprehensive comparison similarity acquisition of the first grayscale image, and divides the historical palmprint image into types through interval comparison, without marking the first grayscale image by artificial means, realizing unsupervised machine learning of the feature palmprint recognition model.

[0193] An image recognition model is created through an existing artificial intelligence platform, the image recognition model is trained using the palmprint training image marking data, and a feature palmprint recognition model is obtained;

[0194] Specifically as follows:

[0195] The palmprint training image marking data is divided into a palmprint image training set and a palmprint image test set according to an image training test ratio;

[0196] It should be noted here that:

[0197] In the present application, the image training test ratio is specifically set to 7:3, that is, the number ratio of medical palmprint images in the palmprint image training set and the palmprint image test set is 7:3;

[0198] An image recognition model is created through an existing artificial intelligence platform, the image recognition model is trained using the palmprint image training set, and each palmprint image in the palmprint image training set is trained once on the image recognition model;

[0199] The image recognition model is tested using the palmprint image test set, and the recognition accuracy is obtained, when the recognition accuracy is greater than or equal to the target recognition accuracy, the image recognition model training is completed, and the feature palmprint recognition model is obtained, when the recognition accuracy is less than the target recognition accuracy, the image recognition model is continuously trained using the palmprint image training set, until the recognition accuracy is greater than or equal to the target recognition accuracy.

[0200] It should be noted here that:

[0201] The target recognition accuracy referred to here is specifically set to 95% in the present application.

[0202] The feature palmprint recognition model is used to identify the to-be-identified palmprint image, if the feature palmprint recognition model output result is image comparison pass, the biometric recognition is completed, if the feature palmprint recognition model output result is image comparison fail, the biometric recognition fails;

[0203] It should be noted here that:

[0204] The second biometric identification device referred to here can be an outdoor palmprint identification access control, the biometric palmprint identification performed by the first biometric identification device can be human palmprint identification, if the identification is successful, the access control passes, if the identification is not successful, the access control fails.

[0205] It should be noted here that:

[0206] The following advantages exist here:

[0207] The feature palmprint recognition model belongs to an artificial intelligence model, the feature palmprint recognition model can continue to train the feature palmprint recognition model while identifying the biometric palmprint, and can realize unsupervised machine learning, and the feature palmprint recognition model can improve the recognition accuracy compared with a traditional entry mode palmprint recognition model;

[0208] The existing palmprint recognition model can only compare the real-time acquired palmprint image with the palmprint image pre-existing in the second biometric identification device, and output the recognition result according to the comparison result, here, the historical images identified and saved in a historical period are used to train the artificial intelligence model, as the second biometric identification device is used for a longer time, the historical saved images used for training are more abundant, and the recognition ability of the feature palmprint recognition model is stronger, and the recognition accuracy can be greatly improved.

[0209] In the present application, if the corresponding calculation formula appears, the above calculation formula is all de-dimensioned to calculate the numerical value, the weight coefficient, the proportion coefficient and other coefficients existing in the formula are set to a result value obtained by quantizing the parameters, and the size of the weight coefficient and the proportion coefficient only needs to not affect the proportional relationship between the parameters and the result value.

[0210] The preferred embodiments of the application disclosed above are only to help explain the present application. The preferred embodiments are not intended to be exhaustive or to limit the application to the specific form disclosed. Many modifications and variations are possible in light of this teaching. It is intended that the specification be considered as exemplary only with the factual recitations being intended to be within the scope of the patent and claims as set forth herein.

Claims

1. An artificial intelligence-based multi-modal biometric recognition method, characterized by, Comprise: Step S1: obtain a plurality of outdoor biometric identification area, respectively to each outdoor biometric identification area for inverse light intensity analysis and inverse light angle analysis, according to the analysis result obtains each outdoor biometric identification area corresponding biometric identification influence coefficient, get area inverse light analysis data; Step S2: according to the area inverse light analysis data on outdoor biometric identification area identification mode analysis, according to the analysis result will outdoor biometric identification area is divided into first type biometric identification area and second type biometric identification area, get identification area type division data, and using the first biometric identification device for the first type biometric identification area biological identification; Step S3: according to the identification area type division data on the second type biometric identification area for acquisition, and using the second biometric identification device for the second type biometric identification area in the biological identification biological palm print image analysis, and according to the analysis result biological identification; The step S1, further comprising the following specific steps: Step S11: to the outdoor area for biological identification needs to be obtained, get a plurality of outdoor biometric identification area, and in the sample biometric identification area selected in the obtained outdoor biometric identification area, and set the first biometric identification device and the second biometric identification device in the sample biometric identification area respectively; Step S12: inverse light intensity analysis and inverse light angle analysis are carried out on the sample biometric identification area, and the biological inverse light identification influence coefficient is obtained according to the analysis result; Step S13: the inverse light identification influence coefficient corresponding to each outdoor biometric identification area is obtained, and the area inverse light analysis data is obtained; The step S12, further comprising the following specific steps: Step S121: according to the relative position of the biological body to be identified and the first biometric identification device, the coordinate system is created, and the identification area space coordinate system is obtained; Step S122: according to the identification area space coordinate system, the face inverse light angle analysis is carried out on the biological body to be identified, and the face inverse light angle corresponding to the biological body to be identified is obtained according to the analysis result; Step S123: the light intensity corresponding to the sample biometric identification area at the current time is obtained, and the area real-time light intensity value is obtained; Step S124: the face inverse light angle and the area real-time light intensity value are calculated to obtain the biological inverse light identification influence coefficient corresponding to the sample biometric identification area; The inverse light identification influence coefficient corresponding to the sample biometric identification area is calculated, and the specific formula is as follows: ； Wherein, Nxk is the biological inverse light identification influence coefficient corresponding to the sample biometric identification area, Mjj is the face inverse light angle, Lss is the area real-time light intensity value, and Lck is the reference light intensity value.

2. The multi-modal biometric recognition method based on artificial intelligence according to claim 1, characterized in that, The step S121, further comprising the following specific steps: In the sample biological recognition region, a facial geometric center point corresponding to the to-be-recognized organism is acquired by an image acquisition device to obtain a first position feature point, a straight line is drawn through the first position feature point and horizontally to one side to obtain a first position feature straight line, a straight line is drawn through the first position feature point and vertically to obtain a second position feature straight line, and a straight line is drawn through the first position feature point and horizontally to the other side to obtain a third position feature straight line; The first position feature point is marked as a coordinate origin, the first position feature straight line is marked as a coordinate x-axis, the second position feature straight line is marked as a coordinate y-axis, and the third position feature straight line is marked as a coordinate z-axis to obtain a recognition region space coordinate system.

3. The multi-modal biometric recognition method based on artificial intelligence according to claim 1, characterized in that, The step S122 further includes the following specific steps: Step S1221: performing position analysis on the first biological recognition device according to the recognition region space coordinate system, and obtaining a device line-of-sight vector according to an analysis result; Step S1222: According to the identification area spatial coordinate system, the position of the sunlight is analyzed, and the sunlight vector is obtained according to the analysis result ; Step S1223: obtaining the sunlight vector and the device line-of-sight vector by calculation, the face inverse-lighting angle corresponding to the to-be-identified organism is obtained; An inverse light included angle of the face of the to-be-recognized organism is calculated, and a specific formula is as follows: ; Wherein, Mjj is the face inverse light angle corresponding to the to-be-identified biological, is the sunlight vector, is the device line-of-sight vector.

4. The multi-modal biometric recognition method based on artificial intelligence according to claim 3, characterized in that, The step S1221 further includes the following specific steps: A first horizontal distance value is obtained by acquiring, through the recognition region space coordinate system, a horizontal distance value between a first biological recognition device center point and a facial geometric center point of the organism; A first vertical distance value is obtained by acquiring a vertical distance value between the first biological recognition device center point and the facial geometric center point of the organism; A first device horizontal offset angle is obtained by acquiring an included angle between a horizontal ray passing through the first biological recognition device center point and a vertical plane in front of the face of the organism; A first device pitch angle is obtained by acquiring an included angle between the horizontal ray passing through the first biological recognition device center point and a horizontal plane in which the facial geometric center point of the organism is located. A vector from the first biometric device center point to the biometric face geometric center point is labeled as the device line of sight vector ; Acquisition device line of sight vector The projection length dxz in the coordinate XZ plane: ; Acquisition device line of sight vector The corresponding x-axis component Cx: ; Acquisition device line of sight vector The corresponding z-axis component Cy: ; Acquisition device line of sight vector The corresponding y-axis component Cz: ; wherein d is a first horizontal distance value, h is a first vertical distance value, is a first device pitch angle, is a first device horizontal offset angle; The first horizontal distance value d and the first vertical distance value h are obtained by calculating the length of the device line of sight vector ; and ; By the device line-of-sight vector The corresponding x-axis component Cx, z-axis component Cz, y-axis component Cy, and the module length of the device line-of-sight vector The corresponding x-axis component Cx, z-axis component Cz, y-axis component Cy, and the module length of the device line-of-sight vector The device line-of-sight vector is calculated as follows: 。 5. The multi-modal biometric recognition method based on artificial intelligence according to claim 3, characterized in that, The step S1222 further includes the following specific steps: A solar elevation angle α corresponding to a current moment is obtained, and a solar eastward azimuth angle β relative to a positive east direction at the current moment is obtained; Acquiring a sun ray vector The projection length in the coordinate XZ plane: ; Acquiring a sun ray vector The corresponding x-axis component Sx: ; Acquiring a sun ray vector The corresponding z-axis component Sz: ; Acquiring a sun ray vector The corresponding y-axis component Sy: ; The sun ray vector is then calculated as The projection length on the coordinate XZ plane, the x-axis component, the z-axis component, and the y-axis component can be calculated from the sun ray vector The specific formula is as follows: 。 6. The multi-modal biometric recognition method based on artificial intelligence according to claim 1, characterized in that, The step S2 further includes the following specific steps: Step S21: acquiring region backlight analysis data, and respectively acquiring an inverse light recognition influence coefficient corresponding to each outdoor biological recognition region according to the region backlight analysis data; Step S22: acquiring an inverse light recognition influence coefficient reference interval, if the inverse light recognition influence coefficient is in the inverse light recognition influence coefficient reference interval, an outdoor biological recognition region corresponding to the inverse light recognition influence coefficient is divided into a first type of biological recognition region, if the inverse light recognition influence coefficient is not in the inverse light recognition influence coefficient reference interval, the outdoor biological recognition region corresponding to the inverse light recognition influence coefficient is divided into a second type of biological recognition region, and recognition region type division data is obtained; Step S23: collecting a biological face of the first type of biological recognition region, and performing image consistency comparison on a collection result and an existing biological face image information library, if the comparison passes, the biological recognition is successful, and if the comparison fails, the biological recognition fails.

7. The multi-modal biometric recognition method based on artificial intelligence according to claim 1, characterized in that, The step S3 further includes the following steps: Step S31: acquiring the recognition region type division data, and acquiring the second type of biological recognition region according to the recognition region type division data; Step S32: acquiring a biological palmprint image of the to-be-identified biological in the second-type biological recognition area to obtain a to-be-identified palmprint image; Step S33: acquiring an access palmprint image stored in the second biological recognition device to obtain a plurality of stored palmprint images, creating a palmprint recognition model according to the plurality of stored palmprint images, and obtaining a feature palmprint recognition model; Step S34: identifying the to-be-identified palmprint image using the feature palmprint recognition model, if the feature palmprint recognition model outputs a result of image comparison passing, the biological recognition is completed, and if the feature palmprint recognition model outputs a result of image comparison failing, the biological recognition fails.

8. The multi-modal biometric recognition method based on artificial intelligence according to claim 7, characterized in that, In the step S33, the following steps are further included: A model training image acquisition period is created by taking the time point corresponding to the current time as a period end time point, and palmprint images acquired by the second biological recognition device in the model training image acquisition period are acquired to obtain a plurality of palmprint training images; A sample palmprint training image is randomly selected from the plurality of acquired palmprint training images, and the sample palmprint training image is matched with the stored palmprint image for consistency, and the sample palmprint training image is identified according to the matching result; Specifically as follows: A sample stored palmprint image is randomly selected from the plurality of stored palmprint images, and the palmprint common area in the sample palmprint training image and the sample stored palmprint image is acquired by using an image recognition algorithm, the palmprint common area in the sample palmprint training image is marked as a first palmprint common area, and the palmprint common area in the sample stored palmprint image is marked as a second palmprint common area; The first palmprint common area and the second palmprint common area are respectively subjected to image grayscale processing to obtain a first grayscale image and a second grayscale image; The palmprint covering area in the second grayscale image is marked to obtain a second image palmprint covering area, and the second image palmprint covering area is disassembled into a plurality of pixel points, the pixel depth value of each pixel point is acquired to obtain a plurality of pixel depth values, the maximum pixel depth value is marked as a first palmprint pixel depth reference value, the minimum pixel depth value is marked as a second palmprint pixel depth reference value, and the first palmprint pixel depth reference value and the second palmprint pixel depth reference value are marked as a palmprint pixel point reference interval; The first grayscale image is disassembled into a plurality of pixel points, the pixel depth of each pixel point is acquired to obtain a plurality of pixel depth values, and the pixel point whose pixel depth value is in the palmprint pixel point reference interval is marked as a palmprint region pixel point; In the first grayscale image, the image region occupied by the palmprint region pixel point is marked as a first image palmprint covering area; The first image palmprint covering area and the second image palmprint covering area are overlaid using the first grayscale image, the overlapping area of the first image palmprint covering area and the second image palmprint covering area is named as an image palmprint overlapping area, and the area of the image palmprint overlapping area is acquired to obtain an image palmprint overlapping area value; An area value of the second image palmprint coverage area is obtained, an area value of a palmprint area corresponding to the sample stored palmprint image is obtained, a stored palmprint area value is obtained, a ratio of the area value of the image palmprint overlapping area and the area value of the stored palmprint area is calculated, and a palmprint area similarity between the sample palmprint training image and the sample stored palmprint image is obtained; A palmprint area similarity between the sample palmprint training image and each of the stored palmprint images is obtained respectively, and a palmprint area similarity with the maximum value is marked as a comprehensive comparison similarity corresponding to the sample palmprint training image; A similarity qualified preset interval is obtained, if the comprehensive comparison similarity is in the similarity qualified preset interval, the sample palmprint training image is marked as comparison passed, and if the comprehensive comparison similarity is not in the similarity qualified preset interval, the sample palmprint training image is marked as comparison failed; Each of the palmprint training images is identified and marked respectively, and palmprint training image marking data is obtained; An image recognition model is created, the image recognition model is trained using the palmprint training image marking data, and a feature palmprint recognition model is obtained.

Images