CN114999004A

CN114999004A - Attack recognition method

Info

Publication number: CN114999004A
Application number: CN202210557455.8A
Authority: CN
Inventors: 高志华
Original assignee: Alibaba Cloud Computing Ltd
Current assignee: Alibaba Cloud Computing Ltd
Priority date: 2022-05-20
Filing date: 2022-05-20
Publication date: 2022-09-02

Abstract

One or more embodiments of the present specification provide a method of identifying an attack, the method comprising: under the condition that a living body detection and face recognition request for a target is received, identifying a certificate target contained in the target; detecting whether a human face target exists in the certificate target or not; and under the condition that a human face target exists in the certificate target, judging that the target is certificate attack.

Description

Attack recognition method

Technical Field

One or more embodiments of the present disclosure relate to the field of in-vivo detection technologies, and in particular, to an attack identification method.

Background

In some authentication scenarios, live body detection is also required, and whether the target image is a real face or a fake face attack is distinguished through the real physiological characteristics of an object in the image. The existing in-vivo detection technology widely applied to commercial scenes is easy to be attacked by one type of attack, namely certificate attack, namely, a real person of an attacker and a certificate with an attacked user photo are arranged in the same picture, and because the existing in-vivo detection algorithm only detects and selects one face to carry out in-vivo judgment, when the real person of the attacker and the certificate are in the same picture, the in-vivo detection algorithm can mistakenly select the real face to carry out in-vivo detection, so that the in-vivo detection authentication is successful, and the defense of the in-vivo detection algorithm is broken through; the face recognition algorithm matched with the living body detection algorithm for identity verification may select a user photo on a certificate as a recognition object, so that the user photo is matched with user information, and an attacker passes identity verification. Therefore, how to identify the above kind of attacks is a problem to be solved urgently.

In the prior art, global face recognition is mostly performed on a target, when a plurality of face targets exist in the target, certificate attacks are recognized by sequentially performing live body detection on each face target, although the method can defend against the above types of attacks, the method is easily interfered by other images in the environment, the attack types cannot be correctly judged, and live body detection needs to be performed on each face target, so that the attack recognition efficiency is low.

Disclosure of Invention

In view of this, one or more embodiments of the present specification provide a method of identifying an attack.

To achieve the above object, one or more embodiments of the present disclosure provide the following technical solutions:

according to a first aspect of one or more embodiments of the present specification, there is provided a method of identifying an attack, comprising:

under the condition that a living body detection and face recognition request for a target is received, identifying a certificate target contained in the target;

detecting whether a human face target exists in the certificate target or not;

and under the condition that a human face target exists in the certificate target, judging that the target is certificate attack.

According to a second aspect of one or more embodiments herein, there is provided a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method according to the first aspect.

According to a third aspect of one or more embodiments of the present specification, there is provided an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method according to the first aspect when executing the program.

In the technical scheme provided by the specification, by the method for identifying the certificate target in the target and directly judging the target as the certificate attack image when the certificate target comprises the face target, the certificate attack image can be preferentially identified and processed, the certificate identification is carried out first instead of the face identification, the influence of the environment on the identification result can be avoided, and the efficiency and the accuracy of the identification attack are improved.

Drawings

Fig. 1 is a schematic diagram of an architecture for identifying an attack apparatus according to an exemplary embodiment of the present specification;

FIG. 2 is a flowchart illustrating a method for identifying attacks according to an exemplary embodiment of the present disclosure;

FIG. 3 is a schematic diagram of a target image provided by an exemplary embodiment of the present description;

FIG. 4 is a schematic illustration of another target image provided by an exemplary embodiment of the present description;

fig. 5 is a schematic structural diagram of an electronic device according to an exemplary embodiment of the present disclosure;

fig. 6 is a schematic diagram of an attack recognition apparatus according to an exemplary embodiment of the present disclosure.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. The following description refers to the accompanying drawings in which the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with one or more embodiments of the present specification. Rather, they are merely examples of apparatus and methods consistent with certain aspects of one or more embodiments of the specification, as detailed in the claims which follow.

It should be noted that: in other embodiments, the steps of the corresponding methods are not necessarily performed in the order shown and described herein. In some other embodiments, the methods may include more or fewer steps than those described herein. Moreover, a single step described in this specification may be broken down into multiple steps for description in other embodiments; multiple steps described in this specification may be combined into a single step in other embodiments.

Liveness detection techniques are techniques that determine the true physiological characteristics of a subject in some authentication scenarios. The existing in vivo detection technologies mainly include two categories: one is a matching type living body detection technology, which refers to that in the process of identity verification, a user is matched to carry out combined actions such as blinking, mouth opening, head shaking, head nodding and the like, and the technologies such as human face key point positioning, human face tracking and the like are used for verifying whether the user is a real living body; the other is a silent type living body detection technology, and with the support of the technology, the user does not need to perform the above actions, and the silent type living body detection technology can directly screen whether the image is a living body.

The live body detection technology utilizes a live body detection algorithm to distinguish whether a captured image is a real face or a fake face attack. Various faces displayed by means of other media can be judged as face attacks, such as printed paper photos, face images or videos displayed on a display screen of an electronic product, silica gel masks, three-dimensional 3D (three-dimensional) portrait, photos or videos acquired by an AI (artificial intelligence) face changing technology and the like, and all belong to the scope of face attacks. However, since a certificate with a photo contains both identity information of the owner of the certificate and the photo of the owner of the certificate, such a certificate is often used as a tool for attacking the liveness detection technique and breaking the algorithm defense of the liveness detection technique.

Whether the silent type living body detection technology or the matching type living body detection technology is adopted, the living body detection technology only selects one face to carry out the living body detection. As long as the certificate is held by a real person, or the certificate and the real person are simultaneously positioned in a target image to be subjected to in vivo detection in other forms, because the in vivo detection algorithm only selects one face in the image to carry out in vivo detection, the in vivo detection algorithm detects the face of the real person as a detection target with a certain probability, so that the image is judged to be the in vivo image, and the image breaks through the defense of the in vivo detection algorithm through in vivo detection. The living body detection technology has the capability of identifying the forged face, so the living body detection technology is generally matched with the face identification technology for use, whether the image is attacked by the forged face is detected through the living body detection technology, and whether the face in the image is matched with the user information is verified through the face identification technology, so that double protection is realized. Once the defense of the living body detection algorithm is broken through, the face target in the certificate still exists in the image, so that when face recognition is carried out, the face target on the certificate can be directly used for comparison with user information stored in the database, the image is detected through the face recognition, and adverse consequences such as personal information leakage or property loss of a holder are caused. Therefore, how to identify certificate attacks is an urgent problem to be solved.

In the prior art, most of the methods perform face recognition on the whole image of a target image, and when a plurality of face targets exist in the image, each face target is sequentially subjected to living body detection.

In order to solve the above problems and further improve the identification efficiency for certificate attacks, the present specification proposes a method for identifying attacks. The certificate attack recognition method has the advantages that the certificate target is recognized in advance in the recognition target, and the certificate attack is directly judged when the certificate target comprises the face target, so that certificate attack can be preferentially recognized and processed, the certificate recognition is performed first instead of the face recognition, the influence of the environment on the recognition result can be avoided, and the certificate attack recognition efficiency is improved.

The method of identifying an attack presented in this specification is described below. Referring to fig. 1, fig. 1 is a schematic diagram illustrating an architecture of an attack recognition system according to an exemplary embodiment of the present disclosure.

As shown in fig. 1, the attack recognition system may include: a server 11, a network 12 and at least one terminal 13.

The server 11 may be a physical server including an independent host, or the server 11 may be a virtual server carried by a host cluster. In the operation process, the server 11 may be configured with a device for identifying an attack, which may be implemented in a software and/or hardware manner to provide an attack identification service, perform living body detection and face identification request according to a target sent by the terminal 13, perform attack identification on the target, and determine whether the target is a certificate attack by identifying a certificate target in the target and detecting whether a face target exists in the certificate target.

The terminal 13 is an electronic device that can be used by a user and can initiate a request for living body detection and face recognition of a target. The electronic device may be a mobile phone, a desktop computer, a tablet device, a notebook computer or a handheld computer (PDAs), a wearable device (e.g., smart glasses, VR glasses, etc.), and the like, which is not limited by one or more embodiments of the present disclosure. The terminal 13 may have an image acquisition device for acquiring a target as a living body detection and face recognition object, and the terminal 13 may also receive a target acquired by other equipment through a communication connection, or acquire an image of the target in another manner, which is not limited in this specification.

And the network 12 for interaction between the server 11 and the terminal 13 may include various types of wired or wireless networks.

In an exemplary embodiment of the present specification, the attack recognition system may not include the server 11 and the network 12, and only the terminal 13 may perform offline attack recognition. Specifically, the terminal 13 may further be configured with an attack recognition device, which may be implemented in a software and/or hardware manner to provide an attack recognition service, perform living body detection and face recognition request according to a target sent by the terminal 13, perform attack recognition on the target, and determine whether the target is a certificate attack by identifying a certificate target in the target and detecting whether a face target exists in the certificate target.

In an exemplary embodiment of the present disclosure, the attack recognition system may be an independent system independent from the identity verification system, and is used for performing attack recognition, and of course, the attack recognition system may also be a part of the identity verification system, and performs identity verification through a human face in cooperation with the identity verification system, where the process of identity verification mainly includes live detection and human face recognition.

A method for identifying an attack provided in the present specification is described below with reference to fig. 2. Fig. 2 is a flowchart illustrating a method for identifying an attack according to an exemplary embodiment. As shown in fig. 2, the method may include the steps of:

s201, under the condition that a living body detection and face recognition request for a target is received, a certificate target contained in the target is recognized.

In an exemplary embodiment provided in this specification, it is assumed that when performing authentication, the following two steps are required: firstly, performing living body detection, judging whether a face target in the image is a real person, and entering a face recognition link when the face target in the image is the real person, and comparing the face target in the image with user information stored in a database, thereby completing identity verification.

As shown in fig. 3, it is assumed that a user a holds the certificate of the user B, and when attempting to access the account of the user B by using the terminal device, authentication as described above is required. The certificate is provided with a photo of a user B, and the terminal equipment is provided with an image acquisition device and can acquire images for identity verification.

After the terminal device collects the certificate image of the user B held by the user A, the image is used as a target to initiate a request for living body detection and face recognition of the target, wherein the target is shown as a target 31 in FIG. 3. In the case of receiving a live body detection and face recognition request for the object 31, a credential object included in the object 31, which is a credential of the user B, is recognized, assuming that the credential object is as shown by a credential object 311 in fig. 3.

In an exemplary embodiment of the present specification, the certificate may include any certificate carrying a user photo, for example: one or more of the driver's license, identity card, passport, student's license, etc. are not specifically limited in this specification.

In an exemplary embodiment of the present specification, the terminal device may be a personal computer, a mobile phone, or other terminal device for personal use; other terminal devices for identity verification through human faces can also be used, such as: a gate or an access control system, a vending machine with a face brushing payment function, a ticket checking system of a tourist attraction and the like. The terminal device may have an image acquisition device for acquiring an image and initiating a living body detection and face recognition request for the acquired image, or may receive an image acquired by another device through a communication connection, and initiate a living body detection and face recognition request with the received image as a target image. The Communication connection may include connection via various types of wired or wireless networks, and may also include bluetooth connection, infrared connection, or NFC (Near Field Communication). Of course, the terminal device may also obtain the target image in other manners, and this specification is not limited in particular. In addition, the image may be in the form of a real-time video, a dynamic picture, a static picture, or the like, and different types of images may be selected as objects of identifying attacks according to different application scenarios, which is not limited in this specification.

In an exemplary embodiment of the present specification, the specific process of identifying a credential object contained in an object can be accomplished by the following steps:

first, an image of a target is input into a pre-trained credential target detection model. The certificate object detection model is obtained by pre-training a certificate of a determined type and can identify whether the image has a certificate object of a corresponding type, and after any image is input into the certificate object detection model, whether the image has the certificate object of the certificate type for which the object detection model is directed can be determined according to an output result. The types of the certificates can be one or more.

Then, whether the input image contains the certificate object is determined according to the output result of the certificate object detection model.

For example, assuming that the certificate object 311 included in the image of the object 31 shown in fig. 3 is an identity card, the image of the object 31 shown in fig. 3 is input into a certificate object detection model trained in advance for the identity card, and it is determined that the identity card object-the certificate object 311 is included in the image of the object 31 according to an output result of the certificate object detection model.

After the certificate object included in the image is recognized, the process proceeds to step S202.

S202, detecting whether a human face target exists in the certificate target.

In an exemplary embodiment of the present specification, detecting whether a face object exists in a document object may include the following steps:

first, the position coordinates of the credential object are acquired, for example, as shown in FIG. 3, the position coordinates of the credential object 311 are acquired; then extracting the certificate object 311 according to the position coordinates; and finally detecting whether the extracted certificate object 311 has a human face object. The certificate target is extracted independently, and only the face target is detected, so that the range of an image area for face detection is reduced, and the detection efficiency can be improved.

In an exemplary embodiment of the present specification, the document object may be extracted by a method of intercepting the document object from an image of the object.

In an exemplary embodiment of the present specification, a face detection algorithm is used to perform face detection on a certificate target, and whether a face target exists in the certificate target is determined.

In an exemplary embodiment of the present specification, the position coordinates of the above-mentioned document object may be acquired by the output result of the document object detection model in step S201. And intercepting the target image according to the position coordinates to obtain the image of the corresponding certificate target. For example, after the image of the object 31 is input to the certificate object detection model, the result is the position coordinates corresponding to the certificate object 311, and then the certificate object 311 is cut out from the image of the object 31 according to the position coordinates to obtain the independent image of the cut-out certificate object 311. Then, the face detection is performed on the captured certificate object 311, and since the certificate object 311 includes the face object 3111, it is determined that the face object exists in the certificate object 311.

If the face target exists in the certificate target, the target containing the certificate target can be judged to be certificate attack.

S203, under the condition that the face target exists in the certificate target, judging that the target is certificate attack. For example, in the target 31 shown in fig. 3, the user B tries to impersonate the user a for authentication by holding the certificate of the user a, and in this case, since the target 31 has the certificate target 311 in the image and the certificate target is detected to include the face target 3111, it can be determined that the target 31 is a certificate attack.

Of course, the target may not be a credential attack, e.g., there may not be any credential target in the target.

In this case, in order to avoid other types of attacks, face detection may be performed on the full target image to detect whether a face target exists in the target. And if the human face target exists in the targets, judging whether the human face target is a living body. If the human face target is a living body, the target is not attacked, and under the condition that the human face target is a living body, the human face recognition result generated aiming at the human face target is determined as the human face recognition result of the target.

For example, as shown in fig. 4, it is assumed that the object in fig. 4 is an object 41, there is no certificate object in the object, there is only a face object 411, and the face object 411 is a living body. Then, according to the above steps, when it is recognized that there is no certificate object in the image of the object 41, it can be directly detected whether there is a human face object in the object 41 through a human face detection algorithm, because there is a human face object 411 in the image of the object 41. The live body detection is continued for the face target 411, and since the face target 411 is a live body, the face recognition result generated for the face target can be determined as the face recognition result of the target. In the authentication process, the face recognition result of the face target 411 is used as the face recognition result of the target 41, and if the face target 411 meets the user identity, it can be determined that the target 41 passes the authentication.

However, if the human face target is not a living body, it can be determined that the target is an attack target. For example, if the face target 411 is a printed photograph of the target 41 shown in fig. 4, the face target 411 in the target 41 can be recognized by the face recognition algorithm, but since the face target 411 is a printed photograph and does not have a physiological feature of a living body, the face target 411 can be determined not to be a living body by the living body detection algorithm, and in this case, the target 41 can be determined to be an attack target.

In an exemplary embodiment of the present specification, the above-mentioned liveness detection algorithm may include:

an RGB (color system) image liveness detection algorithm, an infrared image liveness detection algorithm, a 3D Depth (3D Depth information) liveness detection algorithm, and the like.

The RGB image biopsy algorithm adopts a common RGB camera as an image acquisition device, and human image breakouts such as moire fringes, imaging deformity and reflectivity are acquired through analysis, so that identification information required by biopsy is obtained, and the identification accuracy is guaranteed through a multi-dimensional identification basis.

The infrared image biopsy algorithm is added with an infrared camera on the basis of the algorithm capability of the RGB image biopsy technology. Because the infrared image filters the light rays of a specific wave band, the fake human face attack based on screen imaging can be naturally resisted. It is electromagnetic in nature, whether visible or infrared. Object imaging is related to the reflective properties of the material of its surface. The reflection characteristics of attack media such as a real human face, a paper sheet, a screen, a three-dimensional mask and the like are different, so that the imaging effect is different. The difference of the surface material is more obvious in the aspect of infrared wave reflection, so that when a real human face appears in front of the infrared camera and a human face on a screen appears in front of the infrared camera, the pictures captured by the infrared camera have great difference and are very easy to distinguish, and therefore, the infrared image living body detection technology has great advantages in convenient identification of photo activation type attack means.

In addition, the 3D Depth live detection algorithm adopts a Depth camera such as structured light/TOF (Time of Flight), introduces a concept of "Depth information", can obtain 3D data of a face region, and can perform further analysis based on the data, so that counterfeit face attacks of 2D media such as paper photos and screens can be easily distinguished.

In this specification, one or more of different types of live detection algorithms can be selected according to different application scenarios to perform live detection on a human face target in a matching manner.

In another exemplary embodiment of the present specification, a credential object is present in the object, but a human face object is not present on the credential object. In this case, the target is also not a certificate attack. In order to determine whether the target is an attack of another type, it is also necessary to perform a comprehensive face detection on the target to detect whether a face target exists in the target. And if the human face target exists in the targets, judging whether the human face target is a living body. If the human face target is a living body, the target is not attacked, and if the human face target is a living body, the human face recognition result generated aiming at the human face target is determined as the human face recognition result of the target.

In an exemplary embodiment of the present specification, the comprehensive face detection may include performing full-image-wide face detection on an image of a target.

In the two embodiments, if comprehensive face detection is performed on a target to obtain a plurality of face targets, living body detection is performed on the plurality of face targets one by one. Although the above two embodiments also perform full-image face detection on the target, since certificate attacks are already eliminated by the method described above, it is not necessary to perform full-image face detection on each target requiring living body detection and face recognition and perform living body detection on the detected face targets one by one, so that the overall detection efficiency can be improved.

For ease of understanding, a specific exemplary embodiment is described below, assuming that in this embodiment, the following four images are required: the images a, B, C, and D are each subjected to attack recognition. The image A and the image B are both photos shot by a person holding the certificate, wherein the certificate photo in the image A can be clearly observed; and the photo of the certificate in the image B is missing, the image C is a real person, and the image D is a character photo displayed on an electronic screen.

At this time, requests for live body detection and face recognition for the target images A, B, C and D are received, respectively, and the certificate object included in each target image is recognized. And if the certificate target exists in the image, the corresponding output result obtained when the certificate target detection model is used for detecting the target image also comprises the position coordinates of the certificate target in the corresponding image.

Since the certificate object exists in the images a and B, the output results of the images a and B include the position coordinates of the certificate object. The certificate target in the image A is assumed to be a certificate target a; and if the certificate target in the image B is a certificate target B, respectively intercepting the certificate target a and the certificate target B according to the position coordinates of the certificate target a and the certificate target B. And detecting whether a human face target exists in the certificate target a and the certificate target b by using a human face detection algorithm. Because the certificate object a has a human face object, the image A is directly judged to be a certificate attack image.

And because the certificate target B does not have a human face target, the image B is not a certificate attack image, in order to judge whether the image B is an attack of other types, the full-image human face detection is carried out on the image B, the image C without the certificate target and the image D, and whether the detected human face is a living body is judged. And if the images B, C and D all comprise human faces, respectively carrying out living body detection on the human faces in the images by using a living body detection algorithm, and judging the image D as an attack image because the human face in the image D is not a living body. And the images in the images B and C are both living bodies, so that the images B and C can perform face recognition through living body detection, and the face recognition result generated for the face target in the image B can be determined as the face recognition result of the image B, and the face recognition result generated for the face target in the image C can be determined as the face recognition result of the image C.

Fig. 5 is a schematic structural diagram of an electronic device according to an exemplary embodiment of the present specification. Referring to fig. 5, at the hardware level, the apparatus includes a processor 502, an internal bus 504, a network interface 506, a memory 508, and a non-volatile memory 510. Of course it is also possible to include hardware required for other functions. The processor 502 reads the corresponding computer program from the non-volatile memory 510 into the memory 508 and runs it, forming a means of identifying attacks on a logical level. Of course, besides software implementation, the one or more embodiments in this specification do not exclude other implementations, such as logic devices or combinations of software and hardware, and so on, that is, the execution subject of the following processing flow is not limited to each logic unit, and may also be hardware or logic devices.

Corresponding to the embodiment of the method, the present specification further provides an apparatus for identifying an attack, which may include, as shown in fig. 6:

an identification unit 610 configured to identify a credential object included in an object in a case where a living body detection and face recognition request for the object is received;

a detecting unit 620, configured to detect whether a face target exists in the certificate target;

a determining unit 630, configured to determine that the target is a certificate attack if a human face target exists in the certificate target.

Optionally, the identifying unit 610 may be specifically configured to:

inputting the image of the target into a certificate target detection model trained in advance;

and determining whether the image contains the certificate target or not according to the output result of the certificate target detection model.

Optionally, the detecting unit 620 may be specifically configured to:

acquiring the position coordinates of the certificate target;

extracting the certificate target according to the position coordinates;

and detecting whether the extracted certificate targets have human face targets.

Optionally, the detecting unit 620 may be further specifically configured to:

and aiming at the certificate target, carrying out face detection by using a face detection algorithm, and judging whether the certificate target has a face target or not.

Optionally, the apparatus further comprises:

a target face detection unit 640, configured to detect whether a face target exists in the target when the certificate target does not exist in the target or the face target does not exist in the certificate target;

a living body judgment unit 650 for judging whether or not the face target is a living body in a case where the face target exists in the targets;

an attack target determination unit 660 configured to determine that the target is an attack target in a case where the face target is not a living body.

Optionally, the apparatus further comprises:

a result determination unit 670 for determining a face recognition result generated for the face target as a face recognition result of the target in a case where the face target is a living body.

Optionally, the target face detection unit 640 may be specifically configured to:

and aiming at the target, carrying out face detection by using a face detection algorithm, and judging whether a face target exists in the target.

Alternatively, the living body judgment unit 650 may be specifically configured to:

and aiming at the face target, performing living body detection by using a living body detection algorithm, and judging whether the face target is a living body.

The implementation process of the functions and actions of each unit in the above device is specifically described in the implementation process of the corresponding step in the above method, and is not described herein again.

The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. A typical implementation device is a computer, which may take the form of a personal computer, laptop computer, cellular telephone, camera phone, smart phone, personal digital assistant, media player, navigation device, email messaging device, game console, tablet computer, wearable device, or a combination of any of these devices.

In a typical configuration, a computer includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic disk storage, quantum memory, graphene-based storage media or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

In one or more embodiments of the present specification, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

The terminology used in the description of the one or more embodiments is for the purpose of describing the particular embodiments only and is not intended to be limiting of the description of the one or more embodiments. As used in one or more embodiments of the present specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

It should be understood that although the terms first, second, third, etc. may be used herein in one or more embodiments to describe various information, such information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of one or more embodiments herein. The word "if," as used herein, may be interpreted as "at … …" or "when … …" or "in response to a determination," depending on the context.

The above description is only for the purpose of illustrating the preferred embodiments of the one or more embodiments of the present disclosure, and is not intended to limit the scope of the one or more embodiments of the present disclosure, and any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the one or more embodiments of the present disclosure should be included in the scope of the one or more embodiments of the present disclosure.

Claims

1. A method of identifying an attack, comprising:

detecting whether a human face target exists in the certificate target or not;

2. The method of claim 1, wherein said identifying a credential object contained in the object comprises:

3. The method of claim 1, wherein the detecting whether a human face object is present in the document object comprises:

acquiring the position coordinates of the certificate target;

extracting the certificate target according to the position coordinates;

4. The method of claim 1, wherein the detecting whether a human face object is present in the document object comprises:

5. The method of claim 1, further comprising:

detecting whether a human face target exists in the targets under the condition that the certificate target does not exist in the targets or the human face target does not exist in the certificate target;

judging whether the human face target is a living body or not under the condition that the human face target exists in the targets;

and under the condition that the human face target is not a living body, judging that the target is an attack target.

6. The method of claim 5, further comprising:

determining a face recognition result generated for the face target as a face recognition result of the target in a case where the face target is a living body.

7. The method of claim 5, wherein said detecting whether a human face target is present in said targets comprises:

8. The method of claim 5, wherein the determining whether the human face target is a living body comprises:

and aiming at the human face target, performing living body detection by using a living body detection algorithm, and judging whether the human face target is a living body.

9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 8.

10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method according to any of claims 1-8 are implemented when the processor executes the program.