Face attitude angle detection device, method and device and storage medium

Abstract

The invention relates to a face attitude angle detection device, method and device and a storage medium, and relates to the field of face recognition. The device comprises a positioning component, a plurality of supporting components, a fixing component, a data acquisition component and a calculation module, the positioning part is used for providing support for the multiple supporting parts and the fixing part, and the multiple supporting parts, the fixing part and the positioning part are fixedly connected and used for keeping the relative position between the positioning part and the human face fixed; a data acquisition component is arranged on the positioning component and / or each supporting component, and the data acquisition component is used for acquiring pose information of a human face and sending the pose information to the calculation module; and the calculation module is used for determining an attitude angle of the human face based on the pose information. According to the invention, the face attitude angle detection result of the face recognition equipment can be verified.

Description

Technical Field

[0001] The present application relates to the field of detection technology, and in particular to a face posture angle detection device, method, apparatus and storage medium. Background Art

[0002] "Posture angle" refers to the angle of the face relative to the face recognition device in three-dimensional space. It is the rotation angle relative to the three coordinate axes of space in the right-handed Cartesian coordinate system constructed with the tip of the nose as the origin, namely the pitch angle, the yaw angle and the roll angle. "Posture angle" is one of the important indicators for measuring the performance of face recognition image data. Some current industry standards in my country have put forward corresponding requirements for the "posture angle" during face recognition. For example, the "Technical Requirements for Public Security Face Recognition Application Images" (GB / T 35678), "Security Face Recognition Application Systems Part 2: Face Image Data" (GA / T922.2), and "Technical Requirements for Information Technology Biometric Face Recognition Systems" (GB / T 41772-2022) all state that "the horizontal rotation angle of the face should be within ±10°, the pitch angle should be within ±10°, and the tilt angle should be within ±10°." The "Technical Specifications for Security Face Verification Systems at Civil Transport Airports (Trial)" (Civil Aviation Public Issue

[2021] No. 17) states that "face posture: horizontal rotation angle shall not exceed ±35°, pitch angle shall not exceed ±20°, and tilt angle shall not exceed ±45°."

[0003] How to scientifically, objectively and accurately detect the posture angle of a face relative to a facial recognition device has always been a technical challenge faced by the industry. Summary of the invention

[0004] Embodiments of the present disclosure provide a face posture angle detection device, method, apparatus and storage medium.

[0005] In a first aspect, an embodiment of the present disclosure provides a face posture angle detection device, comprising: a positioning component, a plurality of supporting components, a fixing component, a data acquisition component and a computing module; the positioning component is used to provide support for the plurality of supporting components and the fixing component, and the plurality of supporting components and the fixing component are fixedly connected to the positioning component to keep the relative position between the positioning component and the face fixed; a data acquisition component is provided on the positioning component and / or each supporting component, and the data acquisition component is used to collect posture information of the face and send the posture information to the computing module; the computing module is used to determine the posture angle of the face based on the posture information.

[0006] In a second aspect, an embodiment of the present disclosure provides a method for detecting a facial posture angle, which is applied to a facial posture angle detection device as described in the first aspect, wherein the facial posture angle detection device is worn on a human head, and the method comprises: obtaining real-time posture information collected by a data acquisition component, wherein the real-time posture information comprises real-time posture information and real-time distance information; determining a facial reference plane according to the real-time distance information; determining a real-time posture angle of the face according to the real-time posture information; determining a reference posture angle based on the facial reference plane; and determining a true value of the facial posture angle according to the real-time posture angle of the face and the reference posture angle.

[0007] In a third aspect, an embodiment of the present disclosure provides a face posture angle detection device, comprising: a data acquisition unit, configured to acquire real-time posture information collected by a data acquisition component, wherein the real-time posture information includes real-time posture information and real-time distance information; a reference plane determination unit, configured to determine a face reference plane according to the real-time distance information; a posture angle determination unit, configured to determine a real-time posture angle of the face according to the real-time posture information; a reference posture angle determination unit, configured to determine a reference posture angle based on the face reference plane; and a posture angle true value determination unit, configured to determine a face posture angle true value according to the real-time posture angle of the face and the reference posture angle.

[0008] In a fourth aspect, an embodiment of the present disclosure provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the face posture angle detection method as described in the first aspect.

[0009] The technical solution disclosed in the present invention is convenient for people to wear and detect the posture angle of the face. It can be used to verify the face posture angle recognition results of the face recognition device, so as to determine whether the face recognition device meets the relevant standards.

[0010] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0012] Figure 1a , Figure 1b It is a front view and a side view of an embodiment of the face posture angle detection device disclosed in the present invention;

[0013] Figure 2a , Figure 2b A front view and a side view of another embodiment of the face posture angle detection device disclosed in the present invention;

[0014] Figure 3 It is a schematic structural diagram of a buckle of a fixing component of the face posture angle detection device disclosed in the present invention;

[0015] Figure 4a , Figure 4b A schematic diagram of two optional structures of the adjustment component of the face posture angle detection device disclosed in the present invention;

[0016] Figure 5 A schematic diagram of an application scenario of the face posture angle detection device disclosed in the present invention;

[0017] Figure 6 A schematic diagram of a process of an embodiment of face posture angle detection disclosed in the present invention;

[0018] Figure 7 A schematic diagram of a flow chart of another embodiment of face posture angle detection disclosed in the present invention;

[0019] Figure 8 It is a structural schematic diagram of an embodiment of the face posture angle detection device disclosed in the present invention. DETAILED DESCRIPTION

[0020] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present disclosure belongs.

[0021] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present disclosure. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0022] In the absence of conflict, the embodiments in the present disclosure and the features in the embodiments may be combined with each other.

[0023] In order to make the technical solutions and advantages of the present disclosure more clearly understood, the present disclosure is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0024] Figure 1a and Figure 1b FIG. 2 shows a front view and a side view of an embodiment of a face posture angle detection device according to the present disclosure. Figure 1a and Figure 1bAs shown, the face posture angle detection device 100 of this embodiment may include a positioning component 101, a plurality of support components 102 and a fixing component 103. Among them, the positioning component 101 is used to provide positioning support for the support component 102 and the fixing component 103. The support component 102 and the fixing component 103 can be fixedly connected to the positioning component 101 to keep the relative position between the face posture angle detection device 100 and the face fixed and will not shake. In some specific implementations, the positioning component 101 can be a ring-shaped structure, which is convenient for fixing multiple support components 102. The size of the above-mentioned ring-shaped structure can be slightly larger than the size of the human head to avoid blocking the facial features of the human head. The fixing component 103 can be fixed to the back of the human head, so that the face posture angle detection device can be more stably fixed on the human head.

[0025] A plurality of support components 102 may be fixed on the positioning component 101. Specifically, each support component 102 may be fixed at a different position of the positioning component 101. For example, the support components 102 may be arranged at the upper, lower, left, and right positions of the positioning component 101. The support component 102 may be a strip-shaped structure, one end of which is fixed on the positioning component 101, and the other end is located inside the ring structure formed by the positioning component 101, and the data acquisition component 104 may be arranged on the end. In some specific practices, in order to avoid the blocking of the face by each support component 102, the support component 102 may be an elongated strip-shaped structure, and it may be made of a transparent material.

[0026] The data acquisition component 104 may include multiple types of sensors, such as distance sensors, gyroscope sensors, acceleration sensors, magnetic sensors, etc. In some specific practices, the distance sensor in the above-mentioned data acquisition component 104 may be set at one end of the support component 102. When in use, the lower surface of the end of the support component 102 where the distance sensor is set may be attached to the surface of the face. Specifically, the data acquisition component 104 may include multiple distance sensors for measuring the distance between different positions of the face and the face recognition device or the face posture angle recognition device. In order to determine the accuracy of the plane of the face, each distance sensor may be set at a position that can characterize the face plane. In this way, the position data of the face can be collected more accurately, so that the posture angle of the face can be calculated more accurately. The above-mentioned distance sensor can be used to measure the distance between the face and the screen of the face recognition device, the gyroscope sensor can measure the head rotation angular velocity in the three coordinate axis directions in the Cartesian coordinate system, the acceleration sensor can measure the head rotation acceleration in the three coordinate axis directions in the Cartesian coordinate system, and the magnetic sensor can measure the head magnetic heading angle in the three coordinate axis directions in the Cartesian coordinate system. It is understandable that in order to avoid the data acquisition component 104 blocking the human face, some components (such as the gyroscope sensor, the acceleration sensor and the magnetic sensor) can be arranged on the positioning component 101. Only the distance sensor is arranged on the supporting component 102, and a small distance sensor can be used to minimize the blocking of the human face by the distance sensor.

[0027] The data acquisition component 104 may send the collected data to the calculation module 105, and the calculation module 105 may determine the posture angle of the face based on the data collected by the data acquisition component 104. In some specific practices, the calculation module 105 may be implemented using a chip, such as an FPGA chip, a single-chip microcomputer, etc. The calculation module 105 may also output the calculated posture angle of the face for viewing by a technician.

[0028] The face posture angle detection device provided by the above-mentioned embodiment of the present disclosure can be fixed on the human head. The face posture information can be collected through the data acquisition component arranged thereon, and the face posture angle can be determined based on the above-mentioned posture information, thereby realizing the detection of the face posture angle.

[0029] Figure 2a and Figure 2b The front view and side view of another embodiment of the face gesture angle detection device disclosed in the present invention are shown. As shown in FIG2 , the face gesture angle detection device 200 in this embodiment may include: a positioning component 201 , a plurality of support components 202 , a fixing component 203 , and an adjustment component 204 disposed on the support component 202 .

[0030] The positioning component 201 is used to provide positioning support for the supporting component 202 and the fixing component 203. In some specific practices, the positioning component 201 can be a ring-shaped structure, specifically an elliptical structure. The area of ​​the ellipse is larger than the area of ​​the human face, so that the human face can be avoided from being blocked. In order to improve the rigidity of the positioning component 201, the positioning component 201 can include a positioning ring sleeve 2011 and a reinforcing rib 2012. The positioning ring sleeve 2011 can be perpendicular to the reinforcing rib 2012, and is used to improve the strength of the positioning ring sleeve 2011, so that it is not easy to bend, and prevent the positioning ring sleeve 2011 from being deformed when worn.

[0031] In some specific practices, in order to ensure that the positioning component 201 does not block the face of the person, the positioning ring sleeve 2011 and the reinforcing rib 2012 can be made of light materials such as transparent acrylic. The advantage of using transparent materials is that they will not block facial details and will not reduce the similarity value of face recognition. The positioning ring sleeve 2011 can be an elliptical ring with a width of about 20mm and a thickness of about 2 to 5mm, and the inner ring area is slightly larger than the face. In actual operation, the inner ring area can also be adjusted to make positioning ring sleeves 2011 of various sizes to adapt to faces of different sizes. The reinforcing rib 2012 can be perpendicular to the positioning ring sleeve 2011, located at the center of the positioning ring sleeve 2011, with a thickness of about 2mm and a height of about 5mm.

[0032] Multiple support components 202 can be respectively arranged at different positions of the positioning component 201. In some specific practices, the support component 202 may include at least 4 strip structures, and the dimensions thereof may be: 5 mm in width, 2 mm in thickness, and 100 mm in length. The above 4 strip structures may be respectively arranged at the upper, lower, left, and right positions of the positioning component 201. Specifically, when the positioning component 201 is an elliptical structure, 2 of the above 4 strip structures may be arranged on the major axis of the ellipse, and the remaining 2 may be arranged on a horizontal line perpendicular to the above major axis. In order to more accurately locate the face plane, the distance sensors arranged on the above 4 strip components may be respectively arranged at the forehead, the lower jaw, and below the cheekbones of the person.

[0033] The fixing member 203 may include a plurality of fixing straps 2031. The fixing straps 2031 may be connected to the positioning member 201 in a variety of ways. For example, the fixing straps 2031 may be fixed to the positioning member 201 by a buckle or a through hole. In some specific practices, the fixing straps 2031 may be detachably connected to the positioning member 201. The fixing straps 2031 may also be connected in a variety of ways. For example, the fixing straps may be fixedly connected by a buckle 2033. The buckle 2033 may be movably connected to the fixing straps 2031, so that the fixing straps 2031 may be adjusted to fit the heads of people of different sizes. Alternatively, the fixing straps 2031 may be provided with through holes, and the fixing bolts 2034 may be connected to the fixing straps by the through holes. In this way, the fixing straps 2031 may be adjusted according to the head sizes of different people, so that the space formed by the fixing straps 2031 may be more suitable for the heads of people.

[0034] In some specific practices, the fixing component 203 may include three fixing straps 2031, one each for the left, top, and right sides, and may be made of soft plastic or linen. Each fixing strap 2031 is about 300 mm long, about 30 mm wide, and about 2 mm thick. If necessary, fixing straps of various sizes may be processed according to actual conditions. One end of the fixing strap 2031 passes through the connecting hole 2032 and is fixed to the positioning component 201. Figure 3 As shown, the buckle 2033 is a rectangular parallelepiped structure with a size of 40mm×40mm×25mm and is made of plastic. The 40mm×25mm plane has a square through hole with a length of 35mm and a width of 10mm, which is used to allow the fixing belt 2031 to pass through. One of the 40mm×40mm planes has an internal threaded hole with a diameter of M15, which is used to install the fixing bolt 2034 of the fixing belt 2031. The fixing bolt 2034 can be a bolt with a length of 20mm and a diameter of M15, and is made of plastic.

[0035] In some optional ways of this embodiment, the support component 202 is provided with a plurality of through holes 2021, and the positioning component 201 may be provided with protrusions 2013 adapted to the through holes, so as to fix the support component 202 on the positioning component 201. In some specific practices, the size of the through holes may be 2 mm in diameter and 8 mm in center distance. The protrusions 2013 may also be fixed by bolts to enhance the stability of the support component 202.

[0036] The adjusting component 204 may be disposed between the supporting component 202 and the positioning component 201, and is used to adjust the angle of the data acquisition component 205 disposed on the supporting component 202. The adjusting component 204 may be fixed to the positioning component 201 by bolts 2041.

[0037] The data acquisition component 205 may include a distance sensor, a gyroscope sensor, an acceleration sensor, a magnetic sensor, etc. The data acquisition component 205 may send the collected data to the calculation module 206. In some specific practices, the distance sensor in the data acquisition component 205 is arranged at the end of the support component 202, and the gyroscope sensor, the acceleration sensor, and the magnetic sensor may be arranged at other positions outside the face, for example, they may be arranged on the positioning component 201. The distance sensor may be a laser distance sensor, and the adjustment component 204 may adjust the angle of the laser distance sensor so that the lasers emitted by each laser distance sensor can be simultaneously projected onto the screen of the face recognition device.

[0038] When the laser distance measuring sensor is set at one end of the support component 202, the wireless laser distance measuring sensor is large in size and will block a large part of the face. Therefore, in this embodiment, a wired laser distance measuring sensor can be used, and the electronic components of the laser distance measuring sensor are only placed at one end of the support component 202, and the electronic components of the power supply, data processing and other circuits are independently placed in the data acquisition component 205. At this time, the connecting line 2051 of the laser distance measuring sensor can be hidden in the positioning component 201 and the support component 202 for easy use and to ensure the communication stability of the data acquisition component 205.

[0039] The laser ranging sensor is used to measure the distance between a human face and the screen of a face recognition device; the gyroscope sensor is used to measure the angular velocity of a human head; the acceleration sensor is used to measure the acceleration of a human head; and the magnetic sensor is used to measure the magnetic heading angle of a human head.

[0040] The laser distance sensor may be a miniature laser distance sensor, such as the VL53L0X or VL53L1X laser distance sensor known in the market. The volume of the laser distance sensor is only 4.4mm×2.4mm×1mm, which can minimize the occlusion of the face.

[0041] The gyroscope sensor can be used to measure the angular velocity of the head rotation in the three coordinate axis directions in the Cartesian coordinate system, and output the measurement results to the main controller. The gyroscope sensor can use an independent three-axis gyroscope sensor, such as the ADIS16265 produced by EVICES, the TAC-450 fiber optic gyroscope produced by EMCORE, etc., or an integrated gyroscope sensor, such as the ADIS16495 or ADIS16445 produced by ANALOG DEVICES, the STIM300 inertial measurement unit produced by Norwegian Sensonor AS, etc.

[0042] Further, the acceleration sensor 403 is used to measure the acceleration of the head rotation in the three coordinate axis directions in the Cartesian coordinate system, and output the measurement results to the main controller 401. The acceleration sensor 403 can be an independent three-axis acceleration sensor, such as ADXL700 or ADXL703 produced by ANALOG DEVICES, Q-FLEX QA3000, QA2000, QA750 of Honeywell, etc., or an integrated three-axis acceleration sensor, such as ADIS16495 or ADIS16445 produced by ANALOG DEVICES, STIM300 inertial measurement unit produced by Norwegian Sensonor AS, etc.

[0043] Furthermore, the magnetic sensor 404 is used to measure the head magnetic heading angle in the three coordinate axis directions in the Cartesian coordinate system, and output the measurement results to the main controller 401. The magnetic sensor 404 can be an independent magnetic sensor, such as HMC2003, HMC5883L, etc. produced by Honeywell, or an integrated magnetic sensor, such as ADIS16495 or ADIS16445 produced by ANALOG DEVICES, STIM300 inertial measurement unit produced by Norwegian Sensonor AS, etc.

[0044] The calculation module 206 can receive the data collected by the data collection component 205, and perform calculations based on the received data to determine the posture angle of the face. The calculation module 206 can be implemented by FPGA, single-chip microcomputer, etc.

[0045] In some optional ways of this embodiment, the positioning component 201 can be a symmetrical annular structure, and the axis of symmetry can coincide with the midline of the nose of the human face. The above-mentioned multiple support components 202 can further include a horizontal sub-support component and a vertical sub-support component. Correspondingly, the adjustment component can include a first sub-adjustment component and a second sub-adjustment component. Among them, the vertical sub-support component can be arranged on the above-mentioned axis of symmetry, and the horizontal sub-support component can be arranged on a horizontal line perpendicular to the above-mentioned axis of symmetry. The first sub-adjustment component is arranged between the horizontal sub-support component and the fixing component 201, and the second sub-adjustment component is arranged between the vertical sub-support component and the fixing component 201.

[0046] The first sub-adjustment component and the second sub-adjustment component can be right-angled triangle or right-angled trapezoidal gaskets, such as Figure 4a and Figure 4bAs shown. Its width can be the same as the width of the strip structure of the support component 202. In order to adjust the angle of the distance sensor set at one end of the support component 202 and keep the above angle fixed, through holes can be set on the adjustment component 204 and the support component 202, and the adjustment component and the support component are fixed together by bolts. In some specific practices, the size of the above through hole can be 2mm in diameter and 8mm in center distance.

[0047] In some optional implementations of the present embodiment, the vertical sub-support component includes an upper sub-support component and a lower sub-support component, and the upper sub-support component and the lower sub-support component are both arranged on the symmetry axis of the positioning component 201 that coincides with the midline of the nose of the human face. The horizontal sub-support component includes a left sub-support component and a right sub-support component, and the left sub-support component and the right sub-support component are arranged on the same horizontal line perpendicular to the above-mentioned symmetry axis, and the left sub-support component and the right sub-support component are arranged symmetrically about the symmetry axis. It should be noted that the height of the gasket used to adjust the left sub-support component should be the same as the height of the gasket used to adjust the right sub-support component. Similarly, the height of the gasket used to adjust the upper sub-support component should be the same as the height of the gasket used to adjust the lower sub-support component. In this way, it can be ensured that the angle of the distance sensor set on the left sub-support component is consistent with the angle of the distance sensor set on the right sub-support component, and the angle of the distance sensor set on the upper sub-support component is consistent with the angle of the distance sensor set on the lower sub-support component.

[0048] In some optional implementations of the present embodiment, the above-mentioned face posture angle detection device may also include an audio output component not shown in the figure, which is used to output corresponding prompt information according to the distance measured by each distance sensor. Specifically, the audio output component may output the first prompt information when the distance measured by the distance sensor located above and the distance sensor located below are equal. When the distance measured by the distance sensor located on the left and the distance sensor located on the right are equal, the second prompt information is output. Alternatively, voice information may be output all the time during the process of wearing the face posture angle detection device on the face. The face posture angle detection device provided by the above-mentioned embodiments of the present disclosure can be fixed on a person's head, and the posture information of the face can be collected by the data acquisition component arranged thereon, and the posture angle of the face can be determined based on the above-mentioned posture information, thereby realizing the detection of the face posture angle.

[0049] Continue to see Figure 5 , which shows a schematic diagram of an application scenario of the face posture angle detection device disclosed in the present invention. Figure 5In the scenario shown, the tester wears the face posture angle detection device disclosed in the present invention and faces the screen of the face recognition device. Turn the head slightly until the audio player of the face posture angle detection device prompts that the upper and lower distances are the same and the left and right distances are also the same, then freeze the face for 3 seconds so that the calculation module in the face posture angle detection device can record the current posture information, which is the reference posture angle. Then, the head can be slightly turned to determine the real-time posture angle of the face. The calculation module determines the true value of the face posture angle based on the real-time posture angle and the reference posture angle. The above true value of the face posture angle is compared with the posture angle output by the face recognition device to verify whether the posture angle calculation result of the face recognition device meets the requirements.

[0050] See also Figure 6 , which shows a process 600 of an embodiment of a method for detecting facial posture angles according to the present disclosure. Figure 6 As shown, in this embodiment, the face posture angle detection device is worn on the head of a person, and each distance sensor is close to the surface of the face. And by adjusting the fixing components, the face posture angle detection device can be stably worn on the head of a person. The method of this embodiment can be executed by the calculation module of the face posture angle detection device. The method of this embodiment can include the following steps:

[0051] Step 601, obtaining real-time posture information collected by the data collection component.

[0052] In this embodiment, the calculation module can obtain the real-time posture information collected by the data acquisition component. The data acquisition component may include: a gyroscope sensor, an acceleration sensor, a magnetic sensor and a plurality of distance sensors. The real-time posture information includes real-time posture information and real-time distance information. The real-time posture information may be information collected by the gyroscope sensor, the acceleration sensor and the magnetic sensor, and may include the head rotation angular velocity collected by the gyroscope sensor, the head rotation acceleration collected by the acceleration sensor, and the head magnetic heading angle collected by the magnetic sensor. The real-time distance information may include the distance sent by each distance sensor. Each distance sensor, gyroscope sensor, acceleration sensor and magnetic sensor may communicate with the calculation module through a pre-set communication port, thereby transmitting the collected data to the calculation module in real time.

[0053] Step 602: Determine the face reference plane according to the real-time distance information.

[0054] In this embodiment, the face reference plane can be determined based on the real-time distance information. Specifically, it can be determined whether the distance measured by the distance sensor set on the upper sub-support component is equal to the distance measured by the distance sensor set on the lower sub-support component. At the same time, it is determined whether the line connecting the light beam projection points of the two distance sensors passes through the center of the face recognition camera and is perpendicular to the upper and lower edges of the screen. It is also possible to simultaneously determine whether the distance measured by the distance sensor set on the left sub-support component is equal to the distance measured by the distance sensor set on the right sub-support component. When it is detected that both pairs of distances are equal, it can be considered that the plane determined by each distance sensor at this time is the face reference plane.

[0055] Step 603: Determine the real-time posture angle of the face according to the real-time posture information.

[0056] The calculation module can also determine the real-time posture angle of the face based on the real-time posture information. Specifically, the predicted posture angle of the face can be determined based on the conversion matrix between the coordinate system of the face recognition device and the head coordinate system, combined with the data measured by the gyroscope sensor. The predicted posture angle can also be corrected using the measurement results of the acceleration sensor and the magnetic sensor to obtain the corrected posture angle. These can be used as the real-time posture angle.

[0057] Step 604, determining a reference posture angle based on the face reference plane.

[0058] After the face reference plane is determined, the reference posture angle can be determined according to the posture information corresponding to the face reference plane. Specifically, the reference posture angle can be determined using the same principle as step 603.

[0059] Step 605, determining the true value of the face posture angle according to the real-time face posture angle and the reference posture angle.

[0060] After determining the real-time posture angle and the reference posture angle of the face, the true value of the face posture angle can be determined based on the two. Specifically, the real-time posture angle of the face at any time can be subtracted from the reference posture angle, and the obtained value is used as the true value of the face posture angle. Alternatively, the real-time posture angle of the face can be multiplied by a preset coefficient and then subtracted from the reference posture angle, and the obtained value is used as the true value of the face posture angle.

[0061] The face posture angle detection method provided by the above embodiment of the present disclosure can determine the face reference plane through the distance measured by two pairs of distance sensors. Based on the face reference plane, the true value of the face posture angle is determined, which is not only convenient for calculation and high in calculation efficiency, but also can improve the accuracy of the calculation result.

[0062] Figure 7 A process 700 of another embodiment of the method for detecting facial posture angles disclosed herein is shown.

[0063] like Figure 7 As shown, the face posture angle detection method of this embodiment may include the following steps:

[0064] Step 701, obtaining real-time head rotation angular velocity collected by the gyroscope sensor, real-time head rotation acceleration collected by the acceleration sensor, real-time head magnetic heading angle collected by the magnetic sensor, and real-time distance information collected by the distance sensor.

[0065] The real-time position information includes real-time posture information and real-time distance information. The real-time distance information includes a first distance, a second distance, a third distance and a fourth distance measured by a first distance sensor, a second distance sensor, a third distance sensor and a fourth distance sensor respectively.

[0066] Step 702: When it is determined that the first distance is equal to the second distance and the third distance is equal to the fourth distance, a plane located by the first distance sensor, the second distance sensor, the third distance sensor and the fourth distance sensor is used as a face reference plane.

[0067] In this embodiment, the first distance sensor and the second distance sensor can be respectively arranged at the ends of the upper sub-support component and the lower sub-support component. The third distance sensor and the fourth distance sensor can be respectively arranged at the ends of the left sub-support component and the right sub-support component. Specifically, the upper sub-support component can be fixed to the upper part of the positioning component to ensure that the end of the upper sub-support component provided with the first distance sensor is close to the upper part of the face, such as the forehead. The lower sub-support component can be fixed to the lower part of the positioning component to ensure that the end of the lower sub-support component provided with the second distance sensor is close to the lower part of the face, such as the chin. The left sub-support component can be fixed to the left part of the positioning component to ensure that the end of the left sub-support component provided with the third distance sensor is close to the left part of the face, such as below the left cheekbone. The right sub-support component can be fixed to the right part of the positioning component to ensure that the end of the right sub-support component provided with the fourth distance sensor is close to the right part of the face, such as below the right cheekbone. When the first distance is equal to the second distance, it is considered that the nose midline of the face is parallel to the plane of the face recognition device. When the third distance is equal to the fourth distance, it is considered that the horizontal line of the face perpendicular to the nose midline is parallel to the plane of the face recognition device. When the first distance is equal to the second distance and the third distance is also equal to the fourth distance, it is considered that the face plane is parallel to the plane of the face recognition device. The face plane at this time is used as the face reference plane.

[0068] Step 703, determining first posture information according to the real-time head rotation angular velocity; determining second posture information according to the real-time head rotation acceleration and the real-time head magnetic heading angle; fusing the first posture information and the second posture information to determine the real-time posture angle of the face.

[0069] In this embodiment, the first posture information can be determined based on the real-time head rotation angular velocity. Specifically, the conversion matrix between the face recognition device coordinate system and the head coordinate system can be first determined. Then, based on the conversion matrix and the head rotation angular velocity output by the gyroscope sensor, the horizontal rotation angle, the roll angle, and the pitch angle are determined as the first posture information.

[0070] At the same time, the second posture information is determined based on the real-time head rotation acceleration and the real-time head magnetic heading angle. Specifically, the matrix of the acceleration measurement values ​​on the three axes of the head coordinate system can be determined first. Combined with the real-time head rotation acceleration measured by the acceleration sensor, the pitch angle and roll angle can be obtained. The matrix of the acceleration measurement values ​​on the three axes of the head coordinate system can also be determined. Combined with the real-time head magnetic heading angle, the rotation angle can be obtained as the second posture information.

[0071] Finally, the first posture information and the second posture information are fused to determine the real-time posture angle of the face. When fusion is performed, multiple methods can be used to achieve fusion. For example, the first posture information and the second posture information can be weighted. Alternatively, one of the first posture information and the second posture information is selected as the final posture information according to the actual application scenario.

[0072] In some optional implementations of this embodiment, the fusion of the first posture information and the second posture information can be achieved through the following sub-steps:

[0073] Sub-step 7031, establish the state prediction equation, the covariance equation of the prediction error and the state update equation.

[0074] Sub-step 7032, determining the process noise covariance based on the error of the gyro sensor.

[0075] Sub-step 7033, determining the fusion error according to the errors of the gyroscope sensor, the acceleration sensor, and the magnetic sensor.

[0076] Sub-step 7034, determining the measurement noise covariance based on the fusion error.

[0077] Sub-step 7035, substitute the first posture information, the second posture information, the process noise covariance, and the measurement noise covariance into the state prediction equation, the covariance equation of the prediction error, and the state update equation to determine the real-time posture angle of the face.

[0078] In this implementation, the following solutions may be specifically included:

[0079] First, establish the face recognition device coordinate system O-XYZ and the head coordinate system o-xyz. The face recognition device coordinate system is the head coordinate system rotated around the z-axis, x-axis, and y-axis respectively. The result after converting the horizontal rotation angle, theta degrees (roll angle), and pitch angle can be expressed as:

[0080]

[0081] Among them, x s ,y s 、z s is the coordinate in the face recognition device coordinate system, x b ,y b 、z b is the coordinate in the head coordinate system, is the rotation matrix, expressed in Euler angles as follows:

[0082]

[0083] Due to the limitation of human movement function, the pitch angle of the human head relative to the face recognition device cannot reach ±90 degrees, and the maximum posture angle requirements of the face in national and industry standards do not reach 90 degrees, so there is no posture angle singularity problem, that is, the face recognition posture angle can be calculated using Euler angles. When the head moves, the head posture angle can be calculated using formula (2) and the output data of the gyroscope sensor as follows:

[0084] Horizontal rotation angle:

[0085] Roll angle: θ = arctanT 32 (4)

[0086] Pitch Angle:

[0087] Since the dynamic accuracy of the gyroscope sensor is high, but cumulative errors will occur after working for a long time, while the dynamic accuracy of the acceleration sensor and magnetic sensor is poor, but the static angle measurement accuracy is high, the measurement results of the acceleration sensor and magnetic sensor are used to calibrate the gyroscope sensor to improve the solution accuracy.

[0088] The matrix of acceleration measurements on the three axes of the head coordinate system can be expressed as A:

[0089]

[0090] In the above formula, g is the acceleration due to gravity.

[0091] The pitch angle and roll angle measured by the acceleration sensor are calculated as follows:

[0092] Pitch Angle:

[0093] Roll Angle:

[0094] The matrix of the magnetic field intensity measurements on the three axes of the head coordinate system can be expressed as M:

[0095]

[0096] Assume that the local magnetic declination (the angle between true north and magnetic north) is known to be The magnetic heading angle of the head relative to true north is It can be solved as:

[0097]

[0098] According to the Kalman filter principle, the calculated angle of the gyroscope sensor is used as the predicted angle, and the calculated angles of the acceleration sensor and magnetic sensor are used as the real-time measurement angles. The state prediction equation of the head posture angle and the covariance equation of the prediction error are established:

[0099]

[0100] Where: is the optimal predicted value of the head posture angle at time k-1, Using the state at time k-1 for time k Predicted head pose angle, A k is the state transfer matrix, W k is the process noise of the gyroscope sensor; P k,k-1 It is the state P at time k using the state P at time k-1. k-1,k-1 The covariance matrix of the prior prediction error obtained, P k-1,k-1 for The corresponding prediction error covariance matrix, Q k is the process noise covariance matrix of the gyroscope sensor during attitude angle prediction.

[0101] Based on the head posture angle measurement value at time k and the optimal predicted value of the head posture angle at time k-1, the state update equation of the head posture angle can be derived as follows:

[0102]

[0103] P k =(1-G k H k ) k,k-1 (14)

[0104]

[0105] Where: is the updated optimal predicted value of the head posture angle, Z kis the calculated angle of the acceleration sensor and magnetic sensor at time k, G k is the Kalman filter gain at the head posture angle k, P k,k-1 is the covariance of the prior prediction error of the head posture angle at time k, H k is the measurement matrix, P k is the covariance of the posterior prediction error of the head posture angle at time k, R k is the measurement noise covariance of the acceleration sensor and magnetic sensor at time k.

[0106] In the recursive process of Kalman filtering, the noise covariance matrix Q is updated at each step. k and the measurement noise covariance R k , Q k It can be obtained by experimentally measuring the error of the gyroscope sensor. k It can be obtained by calculating the second-order center distance of the errors of the gyroscope sensor, acceleration sensor and magnetic sensor. The process is as follows:

[0107] Assume that the true attitude angle at a certain moment is The attitude angle obtained after the measurement value of the gyroscope sensor is filtered by Kalman filtering can be expressed as the sum of the true attitude angle and the gyroscope sensor error:

[0108]

[0109] The difference between the attitude angles of the gyroscope sensor at two adjacent moments is:

[0110]

[0111] Where: is the error of the gyro sensor attitude angle prediction at time k, is the error of the gyro sensor attitude angle prediction at time k-1.

[0112] The attitude angle obtained after the measurement values ​​of the acceleration sensor and the magnetic sensor are filtered by Kalman filtering can also be expressed as the sum of the true attitude angle and the errors of the acceleration sensor and the magnetic sensor:

[0113]

[0114] The difference between the attitude angles of the acceleration sensor and the magnetic sensor at two adjacent moments is:

[0115]

[0116] Where: is the error of attitude angle prediction of acceleration sensor and magnetic sensor at time k, is the error in the attitude angle prediction of the acceleration sensor and magnetic sensor at time k-1.

[0117] Therefore, the fusion error f of the angle calculated by the gyroscope sensor, acceleration sensor and magnetic sensor at any two adjacent moments is:

[0118]

[0119] Solve the second-order center distance R of the fusion error f k as follows:

[0120]

[0121] Due to δ g and δ a are mutually independent zero-mean white noise sequences, so they satisfy E[δ g ·δ a ]=0. The errors of a single sensor at two adjacent moments are independent of each other, that is, The gyroscope sensor has high stability and measurement accuracy, and its error is much smaller than the error of the acceleration sensor and magnetic sensor and can be ignored. The error variation of the acceleration sensor and magnetic sensor at any two adjacent moments is small, which can be approximated Therefore, substituting the above conclusion into formula (21), the simplified measurement noise covariance of the acceleration sensor and magnetic sensor at time k is obtained as follows:

[0122]

[0123] Step 704, determining a reference posture angle based on the face reference plane.

[0124] In this step, the principle of determining the reference attitude angle is similar to that in step 703 and will not be repeated here.

[0125] Step 705, determining the true value of the face posture angle according to the difference between the real-time face posture angle and the reference posture angle.

[0126] After determining the real-time face pose angle at any time, subtract it from the reference pose angle and use the difference as the true value of the face pose angle.

[0127] After explaining the calculation principle, the attitude angle can be calculated using the following steps:

[0128] ① Calculate the initial attitude angle and initial process noise covariance

[0129] After the calculation module is powered on, it is counted as time k = 0, and the head posture angle (horizontal rotation angle) is calculated using the output data of the gyroscope sensor and formulas (3), (4), and (5) Roll angle θ, pitch angle ψ). The head posture angle measurement value is calculated using the acceleration sensor and magnetic sensor and formulas (7), (8), and (10). The difference between the head posture angle measurement value and the head posture angle is used as the measurement error of the gyroscope to calculate the initial process noise covariance Q of the gyroscope sensor. k , and assume that the initial value of the covariance of the prediction error is P0=1.

[0130] ② Calculate the predicted attitude angle at time k = 1

[0131] At time k=1, the head posture angle is calculated using the output data of the gyroscope sensor and formulas (3), (4), and (5), and the current posture angle is predicted using formula (11): Using the initial process noise covariance Q k , the initial value of the covariance of the prediction error P0 and the formula (12) to calculate the covariance of the prior prediction error P k,k-1 .

[0132] ③Calculate Kalman filter gain

[0133] The measurement noise covariance R is calculated using the output data of the acceleration sensor and the magnetic sensor and formula (22). k , and then use formula (15) to calculate the Kalman filter gain G k .

[0134] ④ Update the covariance of the optimal predicted value of the head posture angle and the posterior prediction error

[0135] The head posture angle measurement value Z is calculated using the output data of the acceleration sensor and the magnetic sensor and formulas (7), (8), and (10). k , and then use formula (13) to calculate the optimal predicted value of the head posture angle (Horizontal rotation angle Roll angle θ, pitch angle ψ), and then use formula (14) to calculate the covariance P of the posterior prediction error of the head posture angle k .

[0136] ⑤Calculate the predicted attitude angle at the moment k>1

[0137] The head posture angle is calculated using the output data of the gyroscope sensor and formulas (3), (4), and (5), and the new posture angle is predicted using formula (11): Using the process noise covariance Q k and the covariance P of the posterior prediction error obtained in step ④ k And formula (12) to calculate the new prior prediction error covariance P k,k-1 .

[0138] ⑥Recursive calculation

[0139] Repeat steps ③ to ⑤.

[0140] The optimal predicted value of the head posture angle calculated in step ④ above (Horizontal rotation angle The roll angle θ and pitch angle ψ) are the calculated real-time face posture angles. Using the same method, the reference posture angle when the face is in the face reference plane is calculated and recorded as Then, the calculated real-time face posture angle is subtracted from the reference posture angle of the face reference plane to obtain the true value of the posture angle of the face after rotation relative to the face recognition device, that is:

[0141]

[0142] The true value of the face roll angle = θ-θ0.

[0143] The true value of the face pitch angle = ψ-ψ0.

[0144] The face posture angle detection method provided by the above-mentioned embodiment of the present disclosure measures the distance between four laser ranging sensors and the screen of the face recognition device so that the face is parallel to the screen of the face recognition device, and can quickly and accurately measure the reference posture angle of the face reference plane, so that the head coordinate system and the face recognition device coordinate system become parallel coordinate systems at this time. Thereafter, any rotation angle of the head can be regarded as a rotation angle relative to this reference posture angle, which simplifies the subsequent calculation process of the face posture angle.

[0145] Further references Figure 8 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of a face posture angle detection device, which corresponds to the method embodiment shown in Figure 1 and can be specifically applied to various electronic devices.

[0146] like Figure 8 As shown, the face posture angle detection device 800 of this embodiment includes: a data acquisition unit 801, a reference plane determination unit 802, a posture angle determination unit 803, a reference posture angle determination unit 804 and a posture angle true value determination unit 805.

[0147] The data acquisition unit 801 is configured to acquire real-time position and posture information acquired by the data acquisition component, wherein the real-time position and posture information includes real-time posture information and real-time distance information.

[0148] The reference plane determining unit 802 is configured to determine the face reference plane according to the real-time distance information.

[0149] The posture angle determination unit 803 is configured to determine the real-time posture angle of the face according to the real-time posture information.

[0150] The reference posture angle determination unit 804 is configured to determine a reference posture angle based on a face reference plane.

[0151] The posture angle true value determination unit 805 is configured to determine the face posture angle true value according to the real-time face posture angle and the reference posture angle.

[0152] In addition, in one embodiment of the present disclosure, a non-transitory computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above method are implemented, for example, including: obtaining real-time posture information collected by a data acquisition component, wherein the real-time posture information includes real-time posture information and real-time distance information; determining a face reference plane based on the real-time distance information; determining a real-time posture angle of the face based on the real-time posture information; determining a reference posture angle based on the face reference plane; and determining a true value of the face posture angle based on the real-time posture angle of the face and the reference posture angle.

[0153] In summary, in the technical solution disclosed in the present invention, multiple laser distance measuring sensors are used to measure the distance between the face and the screen of the face recognition device, so that the face is parallel to the screen of the face recognition device, so that the reference posture angle of the face reference plane can be obtained quickly and accurately. Then, the real-time posture angle of the face at any time is subtracted from the reference posture angle to obtain the true value of the face posture angle, which simplifies the calculation process of the face posture angle.

[0154] The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present disclosure shall be included in the scope of protection of the present disclosure.

Claims

1. A face posture angle detection device, comprising: A positioning component, a plurality of supporting components, a fixing component, a data acquisition component and a computing module; The positioning component is used to provide support for the multiple supporting components and the fixing component, and the multiple supporting components, the fixing component and the positioning component are fixedly connected to keep the relative position between the positioning component and the face fixed; The positioning component and / or each of the supporting components is provided with a data acquisition component, and the data acquisition component is used to collect the posture information of the face and send the posture information to the calculation module; The calculation module is used to determine the posture angle of the face based on the posture information.

2. The face posture angle detection device according to claim 1, wherein: The device further comprises an adjustment component disposed between the support component and the fixing component; The adjusting component is used to adjust the angle of the data acquisition component.

3. The face posture angle detection device according to claim 2, wherein: The supporting component comprises a horizontal sub-supporting component and a vertical sub-supporting component, and the adjusting component comprises a first sub-adjusting component and a second sub-adjusting component; The first sub-adjustment component is disposed between the transverse sub-support component and the positioning component, and the second sub-adjustment component is disposed between the vertical sub-support component and the positioning component.

4. The face posture angle detection device according to claim 3, wherein: The positioning component is a symmetrical structure, and the symmetry axis of the positioning component coincides with the midline of the nose of the human face; The vertical sub-support component comprises an upper sub-support component and a lower sub-support component, and the upper sub-support component and the lower sub-support component are arranged on the symmetry axis; The lateral sub-support component includes a left sub-support component and a right sub-support component, the left sub-support component and the right sub-support component are arranged on the same horizontal line perpendicular to the symmetry axis, and the left sub-support component and the right sub-support component are symmetrically arranged about the symmetry axis.

5. The face posture angle detection device according to claim 4, wherein: The adjustment component is a gasket; The number of gaskets used to adjust the upper sub-support member is the same as the number of gaskets used to adjust the lower sub-support member; The number of gaskets used to adjust the left sub-support component is the same as the number of gaskets used to adjust the right sub-support component.

6. The face posture angle detection device according to claim 5, wherein: The cross section of the gasket is triangular or trapezoidal.

7. The face posture angle detection device according to claim 1, wherein: The supporting component is provided with a plurality of through holes, and the positioning component is provided with protrusions adapted to the through holes, so as to fix the supporting component on the positioning component; The adjusting component is arranged between the positioning component and the supporting component, and is used for adjusting the angle between the supporting component and the plane where the positioning component is located.

8. The face posture angle detection device according to claim 7, wherein: The adjusting component is provided with a through hole having the same size as the through hole on the supporting component, and the adjusting component and the supporting component are fixed through the through hole.

9. The face posture angle detection device according to claim 1, wherein: The fixing member comprises a plurality of fixing straps; The multiple fixing belts are connected through connecting holes and fixed by buckles.

10. The face posture angle detection device according to claim 1, wherein: The positioning component and the supporting component are made of transparent material.

11. The face posture angle detection device according to claim 1, wherein: The data acquisition component includes a plurality of distance sensors, a gyroscope sensor, an acceleration sensor and a magnetic sensor; The multiple distance sensors are used to measure the distance between the face and the screen of the face recognition device, and the distance sensors are laser distance measuring sensors; The gyro sensor is used to measure the angular velocity of the head rotation of the human head; The acceleration sensor is used to measure the rotation acceleration of the human head; The magnetic sensor is used to measure the head magnetic heading angle of a human head.

12. The face posture angle detection device according to claim 11, wherein: The supporting component is a strip-shaped structure, one end of the strip-shaped structure is fixed on the positioning component, and the other end of the strip-shaped structure is provided with a distance sensor; The gyro sensor, the acceleration sensor and the magnetic sensor are arranged at the connection between the strip structure and the positioning component, or are arranged on the positioning component.

13. The face posture angle detection device according to claim 12, wherein: The face posture angle detection device further comprises: an audio output component connected to the calculation module; The calculation module is also used to determine prompt information according to the distances measured by the received distance sensors; The audio output component plays the prompt information in voice.

14. A method for detecting a facial posture angle, applied to the facial posture angle detection device according to any one of claims 1 to 13, wherein the facial posture angle detection device is worn on a person's head, the method comprising: Acquire real-time position and posture information collected by the data collection component, wherein the real-time position and posture information includes real-time posture information and real-time distance information; Determine the face reference plane according to the real-time distance information; Determining a real-time face posture angle according to the real-time posture information; Determining a reference posture angle based on the face reference plane; The true value of the face posture angle is determined according to the real-time face posture angle and the reference posture angle.

15. The method according to claim 14, wherein: The step of acquiring the real-time posture information collected by the data collection component includes: Obtain the real-time head rotation angular velocity collected by the gyroscope sensor, the real-time head rotation acceleration collected by the acceleration sensor, the real-time head magnetic heading angle collected by the magnetic sensor, and the real-time distance information collected by the distance sensor.

16. The method according to claim 15, wherein: The real-time distance information includes a first distance, a second distance, a third distance and a fourth distance measured by the first distance sensor, the second distance sensor, the third distance sensor and the fourth distance sensor respectively; as well as Determining the face reference plane according to the real-time distance information includes: When it is determined that the first distance is equal to the second distance and the third distance is equal to the fourth distance, a plane where the first distance sensor, the second distance sensor, the third distance sensor and the fourth distance sensor are located is used as a face reference plane.

17. The method according to claim 15, wherein: Determining the real-time face posture angle according to the real-time posture information includes: Determining first posture information according to the real-time head rotation angular velocity; Determining second posture information according to the real-time head rotation acceleration and the real-time head magnetic heading angle; The first posture information and the second posture information are integrated to determine a real-time posture angle of the face.

18. The method according to claim 17, wherein: The step of fusing the first posture information and the second posture information to determine the real-time posture angle of the face includes: Establish state prediction equation, prediction error covariance equation and state update equation; Determining a process noise covariance based on an error of the gyro sensor; Determining a fusion error according to errors of the gyro sensor, the acceleration sensor, and the magnetic sensor; Determining a measurement noise covariance based on the fusion error; Substitute the first posture information, the second posture information, the process noise covariance, and the measurement noise covariance into the state prediction equation, the covariance equation of the prediction error, and the state update equation to determine the real-time posture angle of the face.

19. The method according to claim 14, wherein: The determining the true value of the face posture angle according to the real-time face posture angle and the reference posture angle comprises: The true value of the face posture angle is determined according to the difference between the real-time face posture angle and the reference posture angle.

20. A facial posture angle detection device, comprising: A data acquisition unit is configured to acquire real-time position and posture information acquired by the data acquisition component, wherein the real-time position and posture information includes real-time posture information and real-time distance information; A reference plane determination unit is configured to determine a face reference plane according to the real-time distance information; A posture angle determination unit, configured to determine a real-time posture angle of a face according to the real-time posture information; A reference posture angle determination unit, configured to determine a reference posture angle based on the face reference plane; The posture angle true value determination unit is configured to determine the face posture angle true value based on the real-time face posture angle and the reference posture angle.

21. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for detecting facial posture angles as described in any one of claims 14 to 19 is implemented.

Images