Living body recognition method, device, terminal and storage medium based on plane detection

Abstract

The present invention belongs to the field of liveness recognition technology, and in particular, relates to a liveness recognition method, device, terminal and storage medium based on plane detection. The liveness recognition method includes: obtaining a plane image of the detection object; extracting facial feature points on the plane image; determining the overall head posture corresponding to the plane image based on the extracted facial feature points; dividing the extracted facial feature points into multiple local feature groups, and determining the local head posture corresponding to each local feature group; calculating the posture difference between the local head posture and the overall head posture; judging whether the detection object is alive based on the posture difference. The present invention realizes liveness recognition through plane image detection, which is beneficial to improving recognition efficiency and reducing equipment costs.

Description

Technical Field

[0001] The present invention belongs to the technical field of living body recognition, and in particular relates to a living body recognition method, device, terminal and storage medium based on plane detection. Background Art

[0002] At present, liveness recognition is mainly used to determine the real physiological characteristics of the detection object in some identity verification scenarios. For example, in face recognition applications, by detecting combined actions such as blinking, opening the mouth, shaking the head, and nodding, it is verified whether the current detection object is a real living person. It can effectively resist common liveness attack methods such as photos, face swapping, masks, occlusions, and screen reshoots, thereby identifying fraudulent behavior and protecting the interests of users.

[0003] The existing methods for detecting live faces mainly have the following problems: First, the computational time is high. For example, it is necessary to use three-dimensional depth information and the optical flow method to calculate the changes in the non-rigid operation of the face, and the computational process is complicated. Second, it is necessary to use additional biometric identification equipment for coordinated identification, and the equipment cost is high. For example, it is necessary to use additional infrared human body detection equipment to detect the temperature of the detection object, or to use sound collection equipment for joint sound recognition, etc. It can be seen that the existing methods for detecting live faces have the problem of complex computational processes or the need to equip additional identification equipment with high costs. Summary of the invention

[0004] In view of this, the present invention provides a liveness recognition method, device, terminal and storage medium based on plane detection to solve the problems of complex calculation process or high cost of additional recognition equipment in the liveness recognition method in the prior art.

[0005] A first aspect of an embodiment of the present invention provides a living body recognition method based on plane detection, which may include:

[0006] Acquire a plane image of the detection object;

[0007] Extracting facial feature points on the plane image;

[0008] Determine the overall head posture corresponding to the plane image based on the extracted facial feature points;

[0009] Divide the extracted facial feature points into multiple local feature groups, and determine the local head posture corresponding to each local feature group;

[0010] Calculating a posture difference between the local head posture and the overall head posture;

[0011] Based on the posture difference, it is determined whether the detected object is a living body.

[0012] A second aspect of an embodiment of the present invention provides a living body recognition device based on plane detection, which may include:

[0013] A plane image acquisition unit, used to acquire a plane face image of the detected object;

[0014] A feature point extraction unit, used to extract facial feature points of the face plane image;

[0015] An overall posture determination unit, used to determine the overall head posture corresponding to the plane image based on the extracted facial feature points;

[0016] A local posture determination unit, used to divide the extracted facial feature points into a plurality of local feature groups, and determine the local head posture corresponding to each local feature group;

[0017] a posture difference calculation unit, used for calculating the posture difference between the local head posture and the overall head posture;

[0018] A living body judging unit is used to judge whether the detected object is a living body based on the posture difference.

[0019] A third aspect of an embodiment of the present invention provides an identification terminal, comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, it implements the steps of the liveness identification method based on plane detection in the first aspect of the present invention and any possible implementation method of the first aspect.

[0020] The fourth aspect of an embodiment of the present invention provides a computer-readable storage medium, which stores computer-readable instructions. When the computer-readable instructions are executed by a processor, the steps of the liveness recognition method based on plane detection in the first aspect of the present invention and any possible implementation method of the first aspect are implemented.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] The present invention obtains a plane image of the detection object, uses the facial feature points on the plane image to determine the overall head posture and local head posture of the detection object on the plane image, and judges whether the detection object is alive based on the posture difference between the overall head posture and the local head posture. The present invention can solve the problems of the existing living body recognition methods such as complex calculation process or the need to be equipped with additional recognition equipment with high cost; that is, on the one hand, the recognition process of the present invention may not involve three-dimensional depth information, which simplifies the calculation process and is conducive to improving the efficiency of living body recognition; on the other hand, the recognition process of the present invention does not need to introduce additional biometric recognition equipment, which reduces the equipment cost for living body recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0024] Figure 1 is a flow chart of an embodiment of a living body recognition method based on plane detection in an embodiment of the present invention;

[0025] Figure 2 for Figure 1 A flow chart of an embodiment of step S103 in the illustrated embodiment;

[0026] Figure 3 is a flow chart of another embodiment of a living body recognition method based on plane detection in an embodiment of the present invention;

[0027] Figure 4 is a structural diagram of an embodiment of a living body recognition device based on plane detection in an embodiment of the present invention;

[0028] Figure 5 The figure is a schematic block diagram of an identification terminal in an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] See also Figure 1 , is a flow chart of an embodiment of a living body recognition method based on plane detection in an embodiment of the present invention, which may include:

[0031] In step S101, a planar image of the detection object is acquired.

[0032] In the embodiment of the present invention, firstly, a plane image of the detection object is obtained, and the plane image refers to a two-dimensional image, which can be obtained by a terminal equipped with an image sensor. For example, the plane image of the detection object can be obtained by a mobile phone terminal equipped with a camera.

[0033] In step S102, facial feature points on the plane image are extracted.

[0034] In an embodiment of the present invention, a facial feature point refers to a pixel point or a set of pixel points on a plane image that can be used to reflect facial features, and each facial feature point can reflect a feature of the face; specifically, reflected in a plane image, each facial feature point can be a pixel point or a set of multiple adjacent pixel points, such as a pixel block composed of multiple adjacent pixel points.

[0035] In an embodiment of the present invention, the extracted facial feature points may be a plurality of predefined facial feature points. For example, the extracted facial feature points may include pixel points or a set of pixel points corresponding to facial features such as the tip of the nose, the chin, the left corner of the left eye, the right corner of the right eye, the left corner of the mouth, and the right corner of the mouth on a plane image.

[0036] In step S103, the overall head posture corresponding to the planar image is determined based on the extracted facial feature points.

[0037] In the embodiment of the present invention, for a given plane image, the head posture of the person on it may include three angles: pitch (up and down flip, such as lowering the head and looking up), yaw (left and right flip, such as turning the head left or right), and roll (rotation in the plane, such as lowering the head to the left or right). The overall head posture is the orientation of the face in the three-dimensional space, and the orientation may be one of the above three angles in the three-dimensional space.

[0038] In the embodiment of the present invention, different orientations of the face in three-dimensional space are reflected in the plane image, mainly in the different position distributions of each facial feature point. Therefore, the orientation of the face in the plane image in three-dimensional space can be determined based on the position distribution of each extracted facial feature point, that is, the overall head posture corresponding to the plane image in three-dimensional space can be obtained based on the position distribution of the facial feature points on the plane image.

[0039] In one implementation, a neural network for overall head posture estimation can be established, and the plane image with determined head posture can be used as a training sample to train and perform deep learning on the established neural network. The plane image with the head posture to be determined is input into the trained neural network to determine the overall head posture of the plane image.

[0040] In addition, in another implementation, the overall head posture estimation can be achieved by using the appearance features of the face on the plane image through an appearance-based method.

[0041] Optional, such as Figure 2 As shown, an embodiment of the above step S103 may include:

[0042] Step S1031, obtaining the position distribution of the extracted facial feature points on the plane image to obtain a first position distribution.

[0043] Step S1032: adjust the posture of the preset three-dimensional face model, and obtain the position distribution of the projections of the facial feature points on the three-dimensional face model on the two-dimensional plane during the adjustment process to obtain a second position distribution.

[0044] In the embodiment of the present invention, the facial feature points on the three-dimensional face model correspond one-to-one to the facial feature points extracted on the planar image.

[0045] Step S1033: Acquire the spatial posture of the three-dimensional face model when the second position distribution is consistent with the first position distribution, and obtain the target posture.

[0046] Step S1031: Determine the target posture as the overall head posture corresponding to the plane image.

[0047] In an embodiment of the present invention, a standard three-dimensional face model can be set in advance, and the three-dimensional face model can be rotated from a reference direction (for example, a positive direction) to adjust its face orientation, and the position changes of each facial feature point on the three-dimensional face model are monitored. When the position distribution of the projection of the facial feature points on the three-dimensional face model on the two-dimensional plane is consistent or nearly consistent with the position distribution of the facial feature points on the extracted plane image, the azimuth angle rotation information of the three-dimensional face model is obtained. At this time, the azimuth angle rotation information is the head posture of the three-dimensional face model, that is, the overall head posture corresponding to the plane image.

[0048] Specifically, the projection calculation of the facial feature points on the three-dimensional face model on the two-dimensional plane can be achieved using the camera calibration method and the direct linear transformation (DLT) method, that is, the conversion from the world coordinate system (three-dimensional) to the image coordinate system (two-dimensional plane).

[0049] In step S104, the extracted facial feature points are divided into a plurality of local feature groups, and the local head posture corresponding to each local feature group is determined.

[0050] In an embodiment of the present invention, the extracted facial feature points are grouped to obtain multiple local feature groups. For each local feature group, the three-dimensional face model is adjusted in the same manner as above to obtain the local head posture corresponding to the local feature group.

[0051] Optionally, the above step S104 may include:

[0052] Dividing the extracted facial feature points into a plurality of local feature groups with a specified number of facial feature points as a group;

[0053] Calculating the local spatial posture corresponding to each local feature group when the three-dimensional face model is in the target posture;

[0054] The local spatial posture corresponding to each local feature group when the three-dimensional face model is in the target posture is determined as the local head posture corresponding to the local feature group.

[0055] In the embodiment of the present invention, the specified number is an integer not less than 3, and the specified number is less than the number of extracted facial feature points. For example, the specified number may be 3, that is, every three facial feature points serve as a local feature group.

[0056] Exemplarily, the three facial feature points of the tip of the nose, the left corner of the left eye, and the right corner of the right eye are divided into a group to obtain a first local feature group. When the three-dimensional face model is in the above-mentioned target posture, the orientation of the plane corresponding to the three facial feature points of the tip of the nose, the left corner of the left eye, and the right corner of the right eye on the three-dimensional face model is determined as the local head posture corresponding to the first local feature group.

[0057] Similarly, the three facial feature points of the chin, the left corner of the left eye, and the right corner of the right eye are divided into a group to obtain a second local feature group. When the three-dimensional face model is in the above-mentioned target posture, the orientation of the plane where the three facial feature points of the chin, the left corner of the left eye, and the right corner of the right eye of the three-dimensional face model are located is determined as the local head posture corresponding to the second local feature group.

[0058] In step S105, a posture difference between the local head posture and the overall head posture is calculated.

[0059] In the embodiment of the present invention, the posture differences between the local head posture corresponding to each local feature group and the overall head posture of the plane image are counted, and further, whether the detected object is alive can be judged by the statistical posture differences.

[0060] Furthermore, the above step S105 may include:

[0061] Obtaining a posture vector of the three-dimensional face model when it is in the target posture;

[0062] Obtaining a plane normal vector of each local feature group when the three-dimensional face model is in the target posture;

[0063] The angle between each plane normal vector and the posture vector is calculated, wherein the size of the angle represents the size of the posture difference.

[0064] In an embodiment of the present invention, the overall head posture and the local head posture are vectorized and represented. The overall head posture is represented by a posture vector when the three-dimensional face model is in the target posture. The local head posture is represented by a plane normal vector of the plane where each local feature group is located when the three-dimensional face model is in the target posture. The angle between each plane normal vector and the posture vector when the three-dimensional face model is in the target posture is calculated. The posture difference between the local head posture and the overall head posture is represented by the angle. The larger the angle, the greater the posture difference between the local head posture and the overall head posture. The smaller the angle, the smaller the posture difference between the local head posture and the overall head posture.

[0065] In step S106, it is determined whether the detected object is a living body based on the posture difference.

[0066] In an embodiment of the present invention, the posture difference reflects the difference between the local plane and the overall posture of the head. In fact, for a living object, the posture difference usually does not exceed a threshold. Therefore, by comparing the posture difference with the threshold, it can be determined whether the detected object is a living object. For example, when the posture difference between the local head posture corresponding to a local feature group and the overall head posture is greater than the set threshold, it can be determined that the detected object is not a living object.

[0067] Furthermore, the above step 106 may include:

[0068] Counting a first number of local feature groups whose angles with the posture vector are greater than a first preset threshold;

[0069] If the first number is greater than a first specified value, it is determined that the detected object is not a living body;

[0070] If the first number is not greater than the first specified value, the detection object is determined to be a living body.

[0071] In an embodiment of the present invention, in order to improve recognition accuracy, the first number of local feature groups whose angles with the posture vector are greater than a first preset threshold can be counted. When the first number is greater than a first specified value, it is determined that the object of this detection is not a living body. When the first number is not greater than the first specified value, it is determined that the object of this detection is a living body.

[0072] For example, when a tablet computer or display is used for a live attack (the detection object is a person on the tablet computer or display), the head posture direction (overall head posture) of the face will remain basically unchanged as the angle between the display plane and the direction of the camera (device for collecting planar images) changes. This is because the video (image) of the attack avatar ensures that, compared with the natural face, the angle between the local plane (local head posture) and the direction of the camera will be incorrect, that is, the two vectors will have an angle difference (posture difference).

[0073] As another example, assuming that the attack is a handheld photo attack, as mentioned above, the overall posture of the human head remains unchanged, but due to the deformation of the paper and the change in the hand-held tilt angle, these deformations can be detected by the normal vector of the local plane. Therefore, the embodiment of the present invention calculates the angle between the multiple sets of plane normal vectors (local head posture) and the head overall posture vector (overall head posture), and performs liveness determination based on the size of the angle. Since the angle between each set of plane normal vectors corresponding to the real living target and the head overall posture vector will remain within the preset threshold, liveness determination can be performed based on the size of the angle.

[0074] To summarize, the present invention obtains a plane image of the detection object, uses the facial feature points on the plane image to determine the overall head posture and local head posture of the detection object on the plane image, and judges whether the detection object is alive based on the posture difference between the overall head posture and the local head posture. The present invention can solve the problems of complex calculation process or the need to equip with additional recognition equipment with high cost in existing living body recognition methods; that is, on the one hand, the recognition process of the present invention may not involve three-dimensional depth information, simplifies the calculation process, and is conducive to improving the efficiency of living body recognition; on the other hand, the recognition process of the present invention does not need to introduce additional biometric recognition equipment, reducing the equipment cost for living body recognition.

[0075] See also Figure 3 , which is another embodiment of a living body recognition method based on plane detection in an embodiment of the present invention, may include:

[0076] Step S301: Acquire a planar image of the detection object.

[0077] Step S302: extract facial feature points on the plane image.

[0078] Step S303: determining the overall head posture corresponding to the plane image based on the extracted facial feature points.

[0079] Step S304: Divide the extracted facial feature points into a plurality of local feature groups, and determine the local head posture corresponding to each local feature group.

[0080] Step S305: Calculate the posture difference between the local head posture and the overall head posture.

[0081] In the embodiment of the present invention, the above steps S301 to S305 can be specifically referred to in Figure 1 Steps S101 to S105 in the illustrated embodiment are not described in detail here.

[0082] Step S306: outputting an action instruction for instructing the detected object to perform a specified head movement.

[0083] Step S307, monitoring the posture difference change rate when the detection object performs the head movement;

[0084] Step S308: determining whether the detected object is a living body based on the posture difference change rate.

[0085] In the embodiment of the present invention, in order to further improve the recognition accuracy, in actual implementation, the detection object may be required to perform one or more specified posture changes such as left, right, up, down, etc., and since the attack image cannot be fixed, this can improve the effectiveness of the algorithm.

[0086] During the process of the detection object's head changing posture, the angle change rate of each plane normal vector and the head posture vector can also be evaluated in real time. The angle change rate reflects the posture difference change rate of the overall head posture and the local head posture of the detection object when performing the head movement. Liveness recognition can also be determined based on the angle change rate.

[0087] In a specific implementation, a plane image of the detected object after the head movement is obtained, and a posture difference is obtained based on the plane image using the same method as described above. The posture difference is compared with the posture difference corresponding to the plane image before the head movement, and the posture difference change rate can be obtained.

[0088] Furthermore, the above step S308 may include:

[0089] Counting a second number of local feature groups whose posture difference change rate is greater than a second preset threshold;

[0090] If the second number is greater than a second specified value, it is determined that the detected object is not a living body;

[0091] If the second number is not greater than the second specified value, the detection object is determined to be a living body.

[0092] In an embodiment of the present invention, by counting the second number of local feature groups whose posture difference change rate is greater than a second preset threshold, when the second number is greater than a second specified value, it is determined that the object of this detection is not a living body; when the second number is not greater than the second specified value, it is determined that the object of this detection is a living body.

[0093] To summarize, the present invention obtains a plane image of the detection object, uses the facial feature points on the plane image to determine the overall head posture and local head posture of the detection object on the plane image, and judges whether the detection object is alive based on the posture difference between the overall head posture and the local head posture. The present invention can solve the problems of complex calculation process or the need to equip with additional recognition equipment with high cost in existing living body recognition methods; that is, on the one hand, the recognition process of the present invention may not involve three-dimensional depth information, simplifies the calculation process, and is conducive to improving the efficiency of living body recognition; on the other hand, the recognition process of the present invention does not need to introduce additional biometric recognition equipment, reducing the equipment cost for living body recognition.

[0094] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0095] Corresponding to the living body recognition method based on plane detection described in the above embodiment, Figure 4 A structural diagram of an embodiment of a living body recognition device based on plane detection provided by an embodiment of the present invention is shown.

[0096] In this embodiment, the liveness recognition device 4 based on plane detection may include: a plane image acquisition unit 41, a feature point extraction unit 42, an overall posture determination unit 43, a local posture determination unit 44, a posture difference calculation unit 45 and a liveness judgment unit 46.

[0097] A plane image acquisition unit 41 is used to acquire a plane image of the face of the detection object;

[0098] A feature point extraction unit 42, used to extract facial feature points of the facial plane image;

[0099] An overall posture determination unit 43, configured to determine an overall head posture corresponding to the plane image based on the extracted facial feature points;

[0100] A local posture determination unit 44, used to divide the extracted facial feature points into a plurality of local feature groups, and determine the local head posture corresponding to each local feature group;

[0101] a posture difference calculation unit 45, for calculating the posture difference between the local head posture and the overall head posture;

[0102] The living body judging unit 46 is used to judge whether the detected object is a living body based on the posture difference.

[0103] Optionally, the living body recognition device 4 based on plane detection may further include:

[0104] A first acquisition unit, used to acquire the position distribution of the extracted facial feature points on the plane image to obtain a first position distribution;

[0105] A second acquisition unit is used to adjust the posture of a preset three-dimensional face model, and in the adjustment process, obtain the position distribution of the projections of the facial feature points on the three-dimensional face model on the two-dimensional plane to obtain a second position distribution, wherein the facial feature points on the three-dimensional face model correspond one-to-one to the facial feature points extracted on the plane image;

[0106] A third acquisition unit, used to acquire the spatial posture of the three-dimensional face model when the second position distribution is consistent with the first position distribution, to obtain a target posture;

[0107] The overall posture determination unit 43 is specifically configured to determine the target posture as the overall head posture corresponding to the planar image.

[0108] Optionally, the living body recognition device 4 based on plane detection may further include:

[0109] A feature group division unit, used to divide the extracted facial feature points into a plurality of local feature groups with a specified number of facial feature points as a group, wherein the specified number is less than the number of the extracted facial feature points;

[0110] A spatial posture calculation unit, used for calculating the local spatial posture corresponding to each local feature group when the three-dimensional face model is in the target posture;

[0111] The local posture determination unit 44 is specifically used to determine the local spatial posture corresponding to each local feature group when the three-dimensional face model is in the target posture as the local head posture corresponding to the local feature group.

[0112] Optionally, the living body recognition device 4 based on plane detection may further include:

[0113] A first vector acquisition unit, used for acquiring a posture vector of the three-dimensional face model when it is in the target posture;

[0114] A second vector acquisition unit, used for acquiring a plane normal vector of each local feature group when the three-dimensional face model is in the target posture;

[0115] The posture difference calculation unit 45 is specifically used to calculate the angle between each plane normal vector and the posture vector, wherein the size of the angle represents the size of the posture difference.

[0116] Optionally, the living body recognition device 4 based on plane detection may further include:

[0117] A first quantity counting unit, used for counting a first quantity of local feature groups whose angle with the posture vector is greater than a first preset threshold;

[0118] The living body judgment unit 46 is specifically configured to judge that the detection object is not a living body if the first number is greater than a first specified value; and judge that the detection object is a living body if the first number is not greater than the first specified value.

[0119] Optionally, the living body recognition device 4 based on plane detection may further include:

[0120] an action instruction unit, used for outputting an action instruction for instructing the detection object to perform a specified head action;

[0121] a difference monitoring unit, used to monitor the posture difference change rate when the detection object performs the head movement;

[0122] The living body judging unit 46 is further configured to judge whether the detection object is a living body based on the posture difference change rate.

[0123] Optionally, the living body recognition device 4 based on plane detection may further include:

[0124] A second quantity counting unit, used for counting a second quantity of local feature groups whose posture difference change rate is greater than a second preset threshold;

[0125] The living body judgment unit 46 is further specifically configured to judge that the detection object is not a living body if the second number is greater than a second specified value; and judge that the detection object is a living body if the second number is not greater than the second specified value.

[0126] To summarize, the present invention obtains a plane image of the detection object, uses the facial feature points on the plane image to determine the overall head posture and local head posture of the detection object on the plane image, and judges whether the detection object is alive based on the posture difference between the overall head posture and the local head posture. The present invention can solve the problems of complex calculation process or the need to equip with additional recognition equipment with high cost in existing living body recognition methods; that is, on the one hand, the recognition process of the present invention may not involve three-dimensional depth information, simplifies the calculation process, and is conducive to improving the efficiency of living body recognition; on the other hand, the recognition process of the present invention does not need to introduce additional biometric recognition equipment, reducing the equipment cost for living body recognition.

[0127] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices, modules and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0128] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0129] Figure 5 A schematic block diagram of an identification terminal provided by an embodiment of the present invention is shown. For ease of description, only the part related to the embodiment of the present invention is shown.

[0130] In this embodiment, the identification terminal 5 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The identification terminal 5 may include: a processor 50, a memory 51, and a computer-readable instruction 52 stored in the memory 51 and executable on the processor 50, such as a computer-readable instruction for executing the above-mentioned live body identification method based on plane detection. When the processor 50 executes the computer-readable instruction 52, the steps in the above-mentioned live body identification method based on plane detection are implemented, such as Figure 1 Alternatively, when the processor 50 executes the computer-readable instructions 52, the functions of each unit in the above-mentioned device embodiments are realized, for example, Figure 4 The functions of the units 41 to 46 are shown.

[0131] Exemplarily, the computer-readable instructions 52 may be divided into one or more modules / units, which are stored in the memory 51 and executed by the processor 50 to implement the present invention. The one or more modules / units may be a series of computer-readable instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer-readable instructions 52 in the identification terminal 5.

[0132] The processor 50 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0133] The memory 51 may be an internal storage unit of the identification terminal 5, such as a hard disk or memory of the identification terminal 5. The memory 51 may also be an external storage device of the identification terminal 5, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the identification terminal 5. Further, the memory 51 may also include both an internal storage unit of the identification terminal 5 and an external storage device. The memory 51 is used to store the computer-readable instructions and other instructions and data required by the identification terminal 5. The memory 51 may also be used to temporarily store data that has been output or is to be output.

[0134] Each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0135] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of computer-readable instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store computer-readable instructions.

[0136] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A living body recognition method based on plane detection, characterized in that: include: Acquire a plane image of the detection object; Extracting facial feature points on the plane image; Determine the overall head posture corresponding to the plane image based on the extracted facial feature points, the overall head posture being the orientation of the face in three-dimensional space, and determine the orientation of the face in three-dimensional space on the plane image according to the position distribution of the extracted facial feature points on the plane figure; Divide the extracted facial feature points into multiple local feature groups, and determine the local head posture corresponding to each local feature group; Calculating a posture difference between the local head posture and the overall head posture; Determining whether the detected object is a living body based on the posture difference; The calculating the posture difference between the local head posture and the overall head posture comprises: Obtaining the posture vector of the 3D face model when it is in the target posture; Obtaining a plane normal vector of each local feature group when the three-dimensional face model is in the target posture; The angle between each plane normal vector and the posture vector is calculated, wherein the size of the angle represents the size of the posture difference.

2. The method for living body recognition based on plane detection according to claim 1, characterized in that: The step of determining the overall head posture corresponding to the plane image based on the extracted facial feature points comprises: Acquire the position distribution of the extracted facial feature points on the plane image to obtain a first position distribution; Adjusting the posture of a preset three-dimensional face model, and obtaining the position distribution of projections of facial feature points on the three-dimensional face model on a two-dimensional plane during the adjustment process to obtain a second position distribution, wherein the facial feature points on the three-dimensional face model correspond one-to-one to the facial feature points extracted on the plane image; Acquire the spatial posture of the three-dimensional face model when the second position distribution is consistent with the first position distribution, and obtain a target posture; The target posture is determined as the overall head posture corresponding to the planar image.

3. The method for liveness recognition based on plane detection according to claim 2, characterized in that: The step of dividing the extracted facial feature points into a plurality of local feature groups and determining the local head posture corresponding to each local feature group comprises: Dividing the extracted facial feature points into a plurality of local feature groups with a specified number of facial feature points as one group, wherein the specified number is less than the number of the extracted facial feature points; Calculating the local spatial posture corresponding to each local feature group when the three-dimensional face model is in the target posture; The local spatial posture corresponding to each local feature group when the three-dimensional face model is in the target posture is determined as the local head posture corresponding to the local feature group.

4. The method for liveness recognition based on plane detection according to claim 1, characterized in that: The determining whether the detected object is a living body based on the posture difference comprises: Counting a first number of local feature groups whose angles with the posture vector are greater than a first preset threshold; If the first number is greater than a first specified value, it is determined that the detected object is not a living body; If the first number is not greater than the first specified value, the detection object is determined to be a living body.

5. The method for living body recognition based on plane detection according to any one of claims 1 to 3, characterized in that: After calculating the posture difference between the local head posture and the overall head posture, the method further includes: Outputting an action instruction for instructing the detected object to perform a specified head action; Monitoring the posture difference change rate when the detection object performs the head movement; Correspondingly, judging whether the detected object is a living body based on the posture difference is specifically as follows: Whether the detected object is a living body is determined based on the posture difference change rate.

6. The method for living body recognition based on plane detection according to claim 5, characterized in that: The determining whether the detected object is a living body based on the posture difference change rate includes: Counting a second number of local feature groups whose posture difference change rate is greater than a second preset threshold; If the second number is greater than a second specified value, it is determined that the detected object is not a living body; If the second number is not greater than the second specified value, the detection object is determined to be a living body.

7. A living body recognition device based on plane detection, characterized in that: include: A plane image acquisition unit, used to acquire a plane face image of the detected object; A feature point extraction unit, used to extract facial feature points of the face plane image; an overall posture determination unit, configured to determine an overall head posture corresponding to the plane image based on the extracted facial feature points, wherein the overall head posture is the orientation of the face in three-dimensional space, and to determine the orientation of the face in three-dimensional space on the plane image according to the position distribution of the extracted facial feature points on the plane figure; A local posture determination unit, used to divide the extracted facial feature points into a plurality of local feature groups, and determine the local head posture corresponding to each local feature group; a posture difference calculation unit, used for calculating the posture difference between the local head posture and the overall head posture; A living body judging unit, used for judging whether the detected object is a living body based on the posture difference; The calculating the posture difference between the local head posture and the overall head posture comprises: Obtaining the posture vector of the 3D face model when it is in the target posture; Obtaining a plane normal vector of each local feature group when the three-dimensional face model is in the target posture; The angle between each plane normal vector and the posture vector is calculated, wherein the size of the angle represents the size of the posture difference.

8. An identification terminal, comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, characterized in that: When the processor executes the computer-readable instructions, the steps of the living body recognition method based on plane detection as described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium storing computer-readable instructions, characterized in that: When the computer-readable instructions are executed by a processor, the steps of the living body recognition method based on plane detection as described in any one of claims 1 to 6 are implemented.

Images