Pet identity verification method, device, equipment and computer-readable storage medium

Abstract

The present application provides a pet identity verification method, device, equipment and computer-readable storage medium, the method comprising: obtaining a pet picture to be verified, and performing validity verification on the pet picture to be verified; when the pet picture to be verified passes the validity verification, extracting a target pet face image from the pet picture to be verified according to a preset pet face recognition model; and verifying the identity of the pet in the pet picture to be verified according to the target pet face image and the pre-stored pet face image. The present application involves biometrics, which can improve the accuracy of pet identity verification.

Description

Technical Field

[0001] The present application relates to the technical field of biometrics, and in particular to a pet identity verification method, device, equipment and computer-readable storage medium. Background Art

[0002] At present, users can insure their pets with pet insurance. When their pets get sick or have an accident, they can apply for compensation from the insurance company to ensure the safety of their pets and reduce their own financial losses. When a claim is needed, users are required to upload photos of the insured pets. The claims adjuster will compare the uploaded pet photos with the pet photos uploaded when the insurance was previously purchased to determine whether the pet being claimed is the previously insured pet. Only after it is determined that the pet being claimed is the previously insured pet can the claim be made. However, when the claims adjuster manually compares the uploaded pet photos with the pet photos uploaded when the insurance was previously purchased, the comparison workload is large, and the claims adjuster cannot accurately identify forged pet photos and cannot accurately verify the uniqueness of the pet's identity. Therefore, how to perform facial recognition on pets and improve the accuracy of pet identity verification is a problem that needs to be solved urgently. Summary of the invention

[0003] The main purpose of this application is to provide a pet identity verification method, device, equipment and computer-readable storage medium, aiming to perform facial recognition on pets and improve the accuracy of pet identity verification.

[0004] In a first aspect, the present application provides a pet identity verification method, the pet identity verification method comprising the following steps:

[0005] Obtaining a pet picture to be verified, and performing validity verification on the pet picture to be verified;

[0006] When the pet picture to be verified passes the validity check, extracting a target pet face image from the pet picture to be verified according to a preset pet face recognition model;

[0007] The identity of the pet in the to-be-verified pet picture is verified according to the target pet face image and the pre-stored pet face image.

[0008] In a second aspect, the present application further provides a pet identity verification device, the pet identity verification device comprising:

[0009] A verification module, used to obtain the pet picture to be verified and perform validity verification on the pet picture to be verified;

[0010] A pet face recognition module, used for extracting a target pet face image from the pet picture to be verified according to a preset pet face recognition model after the pet picture to be verified passes the validity check;

[0011] The identity verification module is used to verify the identity of the pet in the pet picture to be verified based on the target pet face image and the pre-stored pet face image.

[0012] In a third aspect, the present application also provides a computer device, comprising a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, the steps of the pet identity verification method as described above are implemented.

[0013] In a fourth aspect, the present application also provides a computer-readable storage medium, on which a computer program is stored, wherein when the computer program is executed by a processor, the steps of the pet identity verification method as described above are implemented.

[0014] The present application provides a pet identity verification method, apparatus, device and computer-readable storage medium. The present application verifies the validity of a pet picture to be verified, and after the pet picture to be verified passes the validity verification, accurately extracts a target pet face image from the pet picture to be verified through a pet face recognition model, and then compares the target pet image with the pet face image at the time of insurance, thereby obtaining an accurate pet identity verification result and effectively improving the accuracy of pet identity verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 A flowchart of a pet identity verification method provided in an embodiment of the present application;

[0017] Figure 2 A schematic diagram of key points of a pet on a pet's face in an embodiment of the present application;

[0018] Figure 3 A schematic diagram of a scenario for implementing the pet identity verification method provided in this embodiment;

[0019] Figure 4 A flowchart of another pet identity verification method provided in an embodiment of the present application;

[0020] Figure 5 A schematic block diagram of a pet identity verification device provided in an embodiment of the present application;

[0021] Figure 6A schematic block diagram of another pet identity verification device provided in an embodiment of the present application;

[0022] Figure 7 The present invention is a block diagram showing the structure of a computer device according to an embodiment of the present application.

[0023] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0025] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0026] The embodiment of the present application provides a pet identity verification method, device, computer equipment and computer readable storage medium. The pet identity verification method can be applied to a server, which can be a single server or a server cluster composed of multiple servers.

[0027] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0028] Please refer to Figure 1 , Figure 1 A flowchart of a pet identity verification method provided in an embodiment of the present application.

[0029] like Figure 1 As shown, the pet identity verification method includes steps S101 to S103.

[0030] Step S101: Obtain a pet picture to be verified, and perform validity verification on the pet picture to be verified.

[0031] When making a claim, the user needs to upload the picture of the pet to be verified to the server through the terminal device. Specifically, when a picture upload instruction is detected, the terminal device displays the picture upload page, and imports the photographed pet picture to be verified or the pet picture to be verified obtained locally into the picture upload page. When the user confirms the upload, a pet insurance claim request carrying the pet picture to be verified is generated and sent to the server.

[0032] The server obtains the pet picture to be verified and verifies the validity of the pet picture to be verified. That is, when the server receives the pet insurance claim request sent by the terminal device, it can obtain the pet picture to be verified from the pet insurance claim request in real time and verify the validity of the pet picture to be verified, or when the server receives the pet insurance claim request sent by the terminal device, it writes the pet insurance claim request into the identity verification queue. The server can periodically obtain unprocessed pet insurance claim requests from the identity verification queue, obtain the pet picture to be verified from the pet insurance claim request, and verify the validity of the pet picture to be verified.

[0033] In one embodiment, the method of performing validity check on the pet picture to be verified is specifically as follows: the server determines whether the image clarity of the pet picture to be verified is greater than or equal to a preset clarity threshold; if the image clarity of the pet picture to be verified is greater than or equal to the preset clarity threshold, then further determine whether there is a pet face area in the pet picture to be verified, if there is a pet face area in the pet picture to be verified, then determine that the pet picture to be verified passes the validity check, if there is no pet face area in the pet picture to be verified, then determine that the pet picture to be verified fails the validity check; if the image clarity of the pet picture to be verified is less than the preset clarity threshold, then it can also be determined that the pet picture to be verified fails the validity check. Among them, image clarity is used to characterize the clarity of the image.

[0034] Among them, the server can determine the image clarity of the pet picture to be verified by running a pre-stored clarity detection program. It should be noted that the clarity detection program can be implemented based on Brenner gradient function, Tenengrad gradient function, Laplacian gradient function, SMD (grayscale variance) function and energy gradient function, etc. The clarity threshold can be set based on actual conditions, and this application does not make specific limitations.

[0035] In one embodiment, the pet face region is specifically determined as follows: the server determines whether there is a pet face region in the pet picture to be verified by using a preset pet face detection model, that is, the pet picture to be verified is input into the pet face detection model to determine whether there is a pet face region in the pet picture to be verified. The pet face detection model is obtained through training, that is, firstly collecting pet pictures containing pet face regions and pet pictures not containing pet face regions as sample data sets, and marking pet face key points in the pet pictures containing pet face regions, and then designing a pet face detection model based on a neural network, and training the pet face detection model based on the sample data set until the model converges, thereby obtaining a pet face detection model that can detect pet faces.

[0036] Optionally, the neural network is MTCNN (Multi-task convolutional neural network), and a sample data set is obtained, which is a set of pet pictures marked with pet face key points and a set of pet pictures not containing pet face areas; the MTCNN network is trained based on the sample data set until the MTCNN converges, and a pet face detection model can be obtained.

[0037] Step S102: When the pet picture to be verified passes the validity check, a target pet face image is extracted from the pet picture to be verified according to a preset pet face recognition model.

[0038] When the pet picture to be verified passes the validity check, the server extracts the target pet face image from the pet picture to be verified according to the preset pet face recognition model. When the pet picture to be verified fails the validity check, a reminder message can be sent to the corresponding terminal device to remind the user to re-upload the pet picture to be verified.

[0039] Among them, the pet face recognition model is implemented based on MTCNN (Multi-task convolutional neural network), specifically: by collecting a large number of pet picture sets marked with pet face key points, pet face categories and pet picture sets not containing pet face areas as sample data sets; based on the MTCNN, a pet face recognition model to be trained is designed, and the pet face recognition model to be trained is iteratively trained with the collected sample data sets until the pet face recognition model to be trained converges, so as to obtain the pet face recognition model.

[0040] It should be noted that the pet face recognition model to be trained based on the MTCNN network includes a first convolutional neural network layer (CNN), a second CNN layer and a third CNN layer, and each CNN layer includes an input layer, two convolutional layers, a batch normalization network (BatchNormalization, BN) layer, a pet face classification task, a bounding box regression task and a pet face key point positioning task, wherein the BN layer is used to connect two convolutional layers, and one convolutional layer connects the pet face classification task, the bounding box regression task and the pet face key point positioning task. The bounding box regression task can accurately determine the rectangular box used to select the pet face.

[0041] Optionally, the number of pet key points is nine, including two pet key points for the mouth, six pet key points for the eyes, three pet key points for the left eye and one pet key point for the right eye, and one pet key point for the nose. By adding a BN layer, the network can converge quickly, and the number of key points increases from five to nine, which can extract richer pet facial features and improve the model effect. Please refer to Figure 2 , Figure 2 Schematic diagram of key points of a pet on the face of a pet in an embodiment of the present application. Figure 2 As shown, the pet facial image is marked with 9 pet key points, among which the pet's left eye is marked with pet key point 1, pet key point 2 and pet key point 3, the pet's right eye is marked with pet key point 4, pet key point 5 and pet key point 6, the pet's nose is marked with pet key point 7, and the pet's mouth is marked with pet key point 8 and pet key point 9.

[0042] Step S103: verify the identity of the pet in the to-be-verified pet picture according to the target pet face image and the pre-stored pet face image.

[0043] After extracting the target pet face image, the server verifies the identity of the pet in the pet picture to be verified based on the target pet face image and the pre-stored pet face image. The pre-stored pet face image is the pet face image that the user entered into the server when purchasing pet insurance. Specifically, when the user purchases pet insurance through a terminal device, the user uploads the pet picture to be verified to the server. The server extracts the pet face image from the pet picture to be verified based on the pet face recognition model and stores it in association with the insurance policy number. When the user makes a claim, the pet insurance claim request can be sent to the server through the terminal device, and the server can obtain the associated pre-stored pet face image through the insurance policy number in the pet insurance claim request.

[0044] In one embodiment, the method of pet identity verification is specifically as follows: calculating the similarity between the target pet face image and the pre-stored pet face image; determining whether the similarity between the target pet face image and the pre-stored pet face image is greater than or equal to a preset similarity threshold; if the similarity between the target pet face image and the pre-stored pet face image is greater than or equal to the preset similarity threshold, then determining that the pet in the pet picture to be verified has passed the identity verification; if the similarity between the target pet face image and the pre-stored pet face image is less than the preset similarity threshold, then determining that the pet in the pet picture to be verified has not passed the identity verification.

[0045] It should be noted that the above similarity threshold can be set based on actual conditions, and this application does not make any specific restrictions on this. The similarity is calculated as follows:

[0046]

[0047] Among them, r(X,Y) is the similarity between the pet face image and the target pet face image, (x, y) is any point in the pet face image and the target pet face image, n is the number of points, and the larger the r(X,Y) value is, the more similar the pet face image is to the target pet face image.

[0048] Please refer to Figure 3 , Figure 3 A schematic diagram of a scenario for implementing the pet identity verification method provided in this embodiment, such as Figure 3 As shown, the user sends the pet insurance claim request to the server through the terminal device. The pet insurance claim request carries a picture of the pet to be verified. The server can verify the pet's identity and make claims in real time based on the picture of the pet to be verified in the pet insurance claim request, or it can first store the picture of the pet to be verified in the pet insurance claim request, and then perform pet identity verification and claims based on the picture of the pet to be verified at a scheduled time.

[0049] The pet identity verification method provided in the above embodiment verifies the validity of the pet picture to be verified, and after the pet picture to be verified passes the validity verification, accurately extracts the target pet face image from the pet picture to be verified through the pet face recognition model, and then compares the target pet image with the pet face image at the time of insurance, so as to obtain an accurate pet identity verification result, thereby effectively improving the accuracy of pet identity verification.

[0050] Please refer to Figure 4 , Figure 4 A flowchart of another pet identity verification method provided in an embodiment of the present application.

[0051] like Figure 4 As shown, the pet identity verification method includes steps S201 to 205.

[0052] Step S201: Obtain a pet picture to be verified, and perform validity verification on the pet picture to be verified.

[0053] The server obtains the pet picture to be verified and verifies the validity of the pet picture to be verified. That is, when the server receives the pet insurance claim request sent by the terminal device, it can obtain the pet picture to be verified from the pet insurance claim request in real time and verify the validity of the pet picture to be verified, or when the server receives the pet insurance claim request sent by the terminal device, it writes the pet insurance claim request into the identity verification queue. The server can periodically obtain unprocessed pet insurance claim requests from the identity verification queue, obtain the pet picture to be verified from the pet insurance claim request, and verify the validity of the pet picture to be verified.

[0054] Step S202: When the pet picture to be verified passes the validity check, a preset number of candidate pet face images are extracted from the pet picture to be verified by using a preset pet face recognition model.

[0055] When the pet picture to be verified passes the validity check, the server extracts a preset number of candidate pet face images from the pet picture to be verified through a preset pet face recognition model, that is, by selecting the pet picture to be verified multiple times, and extracting a candidate pet face image each time, multiple candidate pet face images can be extracted after multiple selections. It should be noted that the above preset number can be set based on actual conditions, and this application does not make specific restrictions on this.

[0056] Specifically, a preset number of frame selection point coordinate groups are determined through the pet face recognition model, wherein each frame selection point coordinate group includes four frame selection point coordinates, and the frame selection point coordinates are the coordinates of the frame selection points in the pet picture to be verified; based on the preset number of frame selection point coordinate groups, a preset number of candidate pet face images are extracted from the pet picture to be verified, that is, a candidate pet face image corresponding to a frame selection point coordinate group is extracted from the pet picture to be verified each time, and after extracting a preset number of pet pictures to be verified, a preset number of candidate pet face images can be obtained. Among them, one frame selection point coordinate group corresponds to one candidate pet face image.

[0057] Step S203: Determine the output probability value corresponding to each of the candidate pet face images through the pet face recognition model.

[0058] After extracting a preset number of candidate pet face images, the pet face recognition model is used to determine the output probability value corresponding to each candidate pet face image. It should be noted that the output probability value is the probability value when the pet face image output by the pet face recognition model is a candidate pet face image. The larger the probability value of the candidate pet face image, the greater the output probability.

[0059] Step S204: determining a target pet face image from a preset number of candidate pet face images according to the output probability value corresponding to each of the candidate pet face images.

[0060] After obtaining the output probability value corresponding to each candidate pet face image, the server determines the target pet face image from a preset number of candidate pet face images according to the output probability value corresponding to each candidate pet face image. In one embodiment, the candidate pet face image corresponding to the maximum output probability value is used as the target pet face image.

[0061] In one embodiment, the target pet face image is determined in the following manner: according to the output probability value corresponding to each candidate pet face image, the pet face image to be corrected is determined from a preset number of candidate pet face images, that is, the candidate pet face image corresponding to the maximum output probability value is used as the pet face image to be corrected; the pet face key points of the pet face image to be corrected are obtained, and the pet face image to be corrected is corrected according to the pet face key points of the pet face image to be corrected to obtain the target pet face image. Among them, while extracting a preset number of candidate pet face images from the pet picture to be verified by a preset pet face recognition model, the pet face key points of each candidate pet face image are also obtained by the pet face recognition model. By correcting the tilted pet face image, the uniformity of the pet face image can be guaranteed, which is convenient for subsequent identity verification.

[0062] In one embodiment, the method for correcting the pet face image to be corrected is specifically as follows: determining an affine transformation matrix based on the pet face key points of the pet face image to be corrected and a pre-stored standard pet face image, that is, calculating the rotation angle and the coordinate offset based on the pet face key points of the pet face image to be corrected and the pet face key points in the standard pet face image; determining an affine transformation matrix based on the rotation angle and the coordinate offset; correcting the pet face image to be corrected based on the affine transformation matrix to obtain a target pet face image, that is, calculating the affine transformation coordinates of each pixel in the target pet face image after affine transformation based on the affine transformation matrix, and aligning and correcting the target pet face image based on the affine transformation coordinates of each pixel to obtain a target pet face image.

[0063] Among them, the pet face mark in the pre-stored standard pet face image has nine pet face key points, and the pet face is positive and not tilted. The formula of affine transformation is specifically:

[0064]

[0065] Among them, θ x and θ y is the rotation angle, t x and t yis the coordinate offset, (X sp ,Y sp ) is the coordinate of the pixel point in the target pet face image, (x′, y′) is (X sp ,Y sp ) is the coordinate after affine transformation. The formula can be used to obtain the affine transformation coordinates of each pixel in the target pet face image after affine transformation.

[0066] Step S205: verify the identity of the pet in the to-be-verified pet picture according to the target pet face image and the pre-stored pet face image.

[0067] After extracting the target pet face image, the server verifies the identity of the pet in the pet picture to be verified based on the target pet face image and the pre-stored pet face image. The pre-stored pet face image is the pet face image that the user entered into the server when purchasing pet insurance. Specifically, when the user purchases pet insurance through a terminal device, the user uploads the pet picture to be verified to the server. The server extracts the pet face image from the pet picture to be verified based on the pet face recognition model and stores it in association with the insurance policy number. When the user makes a claim, the pet insurance claim request can be sent to the server through the terminal device, and the server can obtain the associated pre-stored pet face image through the insurance policy number in the pet insurance claim request.

[0068] The pet identity verification method provided in the above embodiment can improve the accuracy of pet face image extraction by performing validity verification on the pet picture to be verified, and after the pet picture to be verified passes the validity verification, extracting multiple candidate face images from the pet picture to be verified through a pet face recognition model, and obtaining a target pet face image from the multiple candidate pet face images, thereby facilitating subsequent accurate verification of the pet identity and further improving the accuracy of pet identity verification.

[0069] Please refer to Figure 5 , Figure 5 A schematic block diagram of a pet identity verification device provided in an embodiment of the present application.

[0070] like Figure 5 As shown, the pet identity verification device 300 includes: a verification module 301, a pet face recognition module 302 and an identity verification module 303.

[0071] The verification module 301 is used to obtain the pet picture to be verified and perform validity verification on the pet picture to be verified.

[0072] The pet face recognition module 302 is used to extract a target pet face image from the pet picture to be verified according to a preset pet face recognition model after the pet picture to be verified passes the validity check.

[0073] The identity verification module 303 is used to verify the identity of the pet in the pet picture to be verified based on the target pet face image and the pre-stored pet face image.

[0074] In one embodiment, the verification module 301 is also used to determine whether the image clarity of the pet picture to be verified is greater than or equal to a preset clarity threshold; if the image clarity of the pet picture to be verified is greater than or equal to the preset clarity threshold, determine whether there is a pet face area in the pet picture to be verified; if there is a pet face area in the pet picture to be verified, determine that the pet picture to be verified has passed the validity check; if there is no pet face area in the pet picture to be verified, determine that the pet picture to be verified has not passed the validity check.

[0075] In one embodiment, the verification module 301 is further used to determine whether there is a pet face area in the pet picture to be verified through a preset pet face detection model.

[0076] In one embodiment, the identity verification module 303 is also used to calculate the similarity between the target pet face image and the pre-stored pet face image; determine whether the similarity between the target pet face image and the pre-stored pet face image is greater than or equal to a preset similarity threshold; if the similarity between the target pet face image and the pre-stored pet face image is greater than or equal to the preset similarity threshold, it is determined that the pet in the pet picture to be verified has passed the identity verification; if the similarity between the target pet face image and the pre-stored pet face image is less than the preset similarity threshold, it is determined that the pet in the pet picture to be verified has not passed the identity verification.

[0077] Please refer to Figure 6 , Figure 6 A schematic block diagram of another pet identity verification device provided in an embodiment of the present application.

[0078] like Figure 6 As shown, the pet identity verification device 400 includes: a verification module 401, a pet face recognition module 402, a probability value determination module 403, an image determination module 404 and an identity verification module 405.

[0079] The verification module 401 is used to obtain the pet picture to be verified and perform validity verification on the pet picture to be verified.

[0080] The pet face recognition module 402 is used to extract a preset number of candidate pet face images from the pet picture to be verified by using a preset pet face recognition model after the pet picture to be verified passes the validity check.

[0081] The probability value determination module 403 is used to determine the output probability value corresponding to each of the candidate pet face images through the pet face recognition model.

[0082] The image determination module 404 is used to determine a target pet face image from a preset number of candidate pet face images according to the output probability value corresponding to each of the candidate pet face images.

[0083] The identity verification module 405 is used to verify the identity of the pet in the pet picture to be verified based on the target pet face image and the pre-stored pet face image.

[0084] In one embodiment, the image determination module 404 is further used to determine a pet face image to be corrected from a preset number of candidate pet face images according to the output probability value corresponding to each of the candidate pet face images; obtain pet face key points of the pet face image to be corrected, and correct the pet face image to be corrected according to the pet face key points of the pet face image to be corrected to obtain a target pet face image.

[0085] In one embodiment, the image determination module 404 is also used to determine the affine transformation matrix based on the pet face key points of the pet face image to be corrected and the pre-stored standard pet face image; according to the affine transformation matrix, the pet face image to be corrected is corrected to obtain the target pet face image.

[0086] It should be noted that technicians in the relevant field can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described device and each module and unit can refer to the corresponding process in the aforementioned pet identity verification method embodiment, and will not be repeated here.

[0087] The apparatus provided in the above embodiment may be implemented in the form of a computer program. The computer program may be Figure 7 Runs on the computer device shown.

[0088] See also Figure 7 , Figure 7 A schematic block diagram of the structure of a computer device provided in an embodiment of the present application. The computer device may be a server.

[0089] like Figure 7 As shown, the computer device includes a processor, a memory, and a network interface connected via a system bus, wherein the memory may include a non-volatile storage medium and an internal memory.

[0090] The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any pet identity verification method.

[0091] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0092] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any pet identity verification method.

[0093] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0094] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be 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. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0095] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0096] Obtaining a pet picture to be verified, and performing validity verification on the pet picture to be verified;

[0097] When the pet picture to be verified passes the validity check, extracting a target pet face image from the pet picture to be verified according to a preset pet face recognition model;

[0098] The identity of the pet in the to-be-verified pet picture is verified according to the target pet face image and the pre-stored pet face image.

[0099] In one embodiment, when the processor verifies the identity of the pet in the to-be-verified pet picture according to the target pet face image and the pre-stored pet face image, it is used to implement:

[0100] Calculating the similarity between the target pet face image and the pre-stored pet face image;

[0101] Determining whether the similarity between the target pet face image and the pre-stored pet face image is greater than or equal to a preset similarity threshold;

[0102] If the similarity between the target pet face image and the pre-stored pet face image is greater than or equal to a preset similarity threshold, it is determined that the pet in the to-be-verified pet picture has passed the identity verification;

[0103] If the similarity between the target pet face image and the pre-stored pet face image is less than a preset similarity threshold, it is determined that the pet in the to-be-verified pet picture has failed the identity verification.

[0104] In one embodiment, after implementing the validity check on the pet picture to be verified, the processor is used to implement:

[0105] Determine whether the image clarity of the pet picture to be verified is greater than or equal to a preset clarity threshold, and if the image clarity of the pet picture to be verified is greater than or equal to the preset clarity threshold, determine whether there is a pet face area in the pet picture to be verified;

[0106] If there is a pet face area in the pet picture to be verified, it is determined that the pet picture to be verified has passed the validity check; if there is no pet face area in the pet picture to be verified, it is determined that the pet picture to be verified has not passed the validity check.

[0107] In one embodiment, when determining whether there is a pet face area in the pet picture to be verified, the processor is used to implement:

[0108] By using a preset pet face detection model, it is determined whether there is a pet face area in the pet picture to be verified.

[0109] In another embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0110] Obtaining a pet picture to be verified, and performing validity verification on the pet picture to be verified;

[0111] When the pet picture to be verified passes the validity check, a preset number of candidate pet face images are extracted from the pet picture to be verified by a preset pet face recognition model;

[0112] Determining the output probability value corresponding to each of the candidate pet face images by using the pet face recognition model;

[0113] Determining a target pet face image from a preset number of candidate pet face images according to the output probability value corresponding to each of the candidate pet face images;

[0114] The identity of the pet in the to-be-verified pet picture is verified according to the target pet face image and the pre-stored pet face image.

[0115] In one embodiment, when the processor determines the target pet face image from a preset number of candidate pet face images according to the output probability value corresponding to each of the candidate pet face images, it is used to implement:

[0116] Determining a pet face image to be corrected from a preset number of candidate pet face images according to the output probability value corresponding to each of the candidate pet face images;

[0117] The pet face key points of the pet face image to be corrected are obtained, and the pet face image to be corrected is corrected according to the pet face key points of the pet face image to be corrected to obtain a target pet face image.

[0118] In one embodiment, when the processor corrects the pet face image to be corrected according to the pet face key points of the pet face image to be corrected to obtain the target pet face image, it is used to implement:

[0119] Determine an affine transformation matrix according to pet face key points of the pet face image to be corrected and a pre-stored standard pet face image;

[0120] The pet face image to be corrected is corrected according to the affine transformation matrix to obtain a target pet face image.

[0121] An embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored. The computer program includes program instructions. The method implemented when the program instructions are executed can refer to the various embodiments of the pet identity verification method of the present application.

[0122] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc., equipped on the computer device.

[0123] It should be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0124] It should also be understood that the term "and / or" used in the specification of this application and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations. It should be noted that, in this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "including a..." does not exclude the presence of other identical elements in the process, method, article or system including the element.

[0125] The serial numbers of the embodiments of the present application are for description only and do not represent the advantages and disadvantages of the embodiments. The above description is only a specific implementation mode of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A pet identity verification method, characterized in that: include: Obtaining a pet picture to be verified, and performing validity verification on the pet picture to be verified; When the pet picture to be verified passes the validity check, a target pet face image is extracted from the pet picture to be verified according to a preset pet face recognition model, wherein the pet face recognition model is designed based on a multi-task convolutional neural network, and the pet face recognition model includes a first convolutional neural network layer, a second convolutional neural network layer, and a third convolutional neural network layer, wherein the first convolutional neural network layer, the second convolutional neural network layer, and the third convolutional neural network layer each include an input layer, two convolutional layers, a batch normalization network layer, a pet face classification task, a frame regression task, and a pet face key point positioning task, wherein the batch normalization network layer is used to connect two convolutional layers, and one convolutional layer connects the pet face classification task, the frame regression task, and the pet face key point positioning task; According to the target pet face image and the pre-stored pet face image, verifying the identity of the pet in the to-be-verified pet picture; The step of extracting a target pet face image from the pet picture to be verified according to a preset pet face recognition model includes: Extracting a preset number of candidate pet face images and pet face key points of each candidate pet face image from the pet picture to be verified by using a preset pet face recognition model; Determining the output probability value corresponding to each of the candidate pet face images by using the pet face recognition model; Determine the candidate pet face image corresponding to the maximum output probability value as the pet face image to be corrected; The pet face image to be corrected is corrected according to the pet face key points of the pet face image to obtain a target pet face image.

2. The pet identity verification method according to claim 1, characterized in that: The step of correcting the pet face image to be corrected according to the pet face key points of the pet face image to be corrected to obtain a target pet face image includes: Determine an affine transformation matrix according to pet face key points of the pet face image to be corrected and a pre-stored standard pet face image; The pet face image to be corrected is corrected according to the affine transformation matrix to obtain a target pet face image.

3. The pet identity verification method according to claim 1 or 2, characterized in that: The step of verifying the identity of the pet in the to-be-verified pet picture according to the target pet face image and the pre-stored pet face image includes: Calculating the similarity between the target pet face image and a pre-stored pet face image; Determining whether the similarity between the target pet face image and the pre-stored pet face image is greater than or equal to a preset similarity threshold; If the similarity between the target pet face image and the pre-stored pet face image is greater than or equal to a preset similarity threshold, it is determined that the pet in the to-be-verified pet picture has passed the identity verification; If the similarity between the target pet face image and the pre-stored pet face image is less than a preset similarity threshold, it is determined that the pet in the to-be-verified pet picture has failed the identity verification.

4. The pet identity verification method according to claim 1 or 2, characterized in that: The validity check of the pet picture to be verified includes: Determine whether the image clarity of the pet picture to be verified is greater than or equal to a preset clarity threshold, and if the image clarity of the pet picture to be verified is greater than or equal to the preset clarity threshold, determine whether there is a pet face area in the pet picture to be verified; If there is a pet face area in the pet picture to be verified, it is determined that the pet picture to be verified has passed the validity check; if there is no pet face area in the pet picture to be verified, it is determined that the pet picture to be verified has not passed the validity check.

5. The pet identity verification method according to claim 4, characterized in that: The determining whether there is a pet face region in the pet picture to be verified includes: By using a preset pet face detection model, it is determined whether there is a pet face area in the pet picture to be verified.

6. A pet identity verification device, characterized in that: The pet identity verification device comprises: A verification module, used to obtain the pet picture to be verified and perform validity verification on the pet picture to be verified; A pet face recognition module, used for extracting a target pet face image from the pet picture to be verified according to a preset pet face recognition model after the pet picture to be verified passes the validity check, wherein the pet face recognition model is designed based on a multi-task convolutional neural network, and the pet face recognition model comprises a first convolutional neural network layer, a second convolutional neural network layer and a third convolutional neural network layer, wherein the first convolutional neural network layer, the second convolutional neural network layer and the third convolutional neural network layer each comprise an input layer, two convolutional layers, a batch normalization network layer, a pet face classification task, a frame regression task and a pet face key point positioning task, wherein the batch normalization network layer is used to connect two convolutional layers, and one convolutional layer connects the pet face classification task, the frame regression task and the pet face key point positioning task; An identity verification module, used to verify the identity of the pet in the pet picture to be verified based on the target pet face image and the pre-stored pet face image; The pet face recognition module is further used to extract a preset number of candidate pet face images and pet face key points of each candidate pet face image from the pet picture to be verified through a preset pet face recognition model; A probability value determination module, used to determine the output probability value corresponding to each of the candidate pet face images through the pet face recognition model; An image determination module, used to determine the candidate pet face image corresponding to the maximum output probability value as the pet face image to be corrected; The image determination module is further used to correct the pet face image to be corrected according to the pet face key points of the pet face image to be corrected to obtain a target pet face image.

7. A computer device, characterized in that: The computer device includes a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, the steps of the pet identity verification method as described in any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of the pet identity verification method according to any one of claims 1 to 5 are implemented.

Images