Risk identification method and device, equipment and medium

Abstract

The embodiment of the invention discloses a risk identification method and device, equipment and a medium. The scheme comprises the following steps: acquiring to-be-identified image data provided by a terminal; performing image conversion processing on three-dimensional image data included in the to-be-recognized image data to obtain converted two-dimensional image data; judging whether the converted two-dimensional image data is matched with the two-dimensional image data contained in the image data to be recognized; and if the converted two-dimensional image data is not matched with the two-dimensional image data included in the to-be-identified image data, determining that the to-be-identified image data has a risk.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, apparatus, device and medium for risk identification. Background Art

[0002] Image recognition is an important technical field of artificial intelligence. With the widespread application of face recognition and fingerprint recognition technologies in mobile payment, privacy protection and identity authentication, risk identification of image data is particularly important.

[0003] Based on this, how to improve the accuracy of risk identification has become a technical problem that needs to be solved urgently. Summary of the invention

[0004] The embodiments of this specification provide a method, apparatus, device and medium for risk identification to solve the problem of inaccurate risk identification in existing risk identification methods.

[0005] To solve the above technical problems, the embodiments of this specification are implemented as follows: A risk identification method provided in an embodiment of this specification includes: Obtain image data to be identified provided by the terminal; Performing image conversion processing on the three-dimensional image data contained in the image data to be identified to obtain converted two-dimensional image data; Determining whether the converted two-dimensional image data matches the two-dimensional image data included in the image data to be identified; If the converted two-dimensional image data does not match the two-dimensional image data included in the image data to be identified, it is determined that the image data to be identified has a risk.

[0006] An embodiment of this specification provides a risk identification device, including: An acquisition module, used for acquiring image data to be recognized provided by a terminal; A conversion module, used for performing image conversion processing on the three-dimensional image data contained in the image data to be identified to obtain converted two-dimensional image data; A first judgment module, used to judge whether the converted two-dimensional image data matches the two-dimensional image data included in the image data to be recognized; The first determination module is configured to determine that the image data to be identified has a risk if the converted two-dimensional image data does not match the two-dimensional image data included in the image data to be identified.

[0007] A risk identification device provided in an embodiment of this specification includes: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: Obtain image data to be identified provided by the terminal; Performing image conversion processing on the three-dimensional image data contained in the image data to be identified to obtain converted two-dimensional image data; Determining whether the converted two-dimensional image data matches the two-dimensional image data included in the image data to be identified; If the converted two-dimensional image data does not match the two-dimensional image data included in the image data to be identified, it is determined that the image data to be identified has a risk.

[0008] An embodiment of the present specification provides a computer-readable medium on which computer-readable instructions are stored. The computer-readable instructions can be executed by a processor to implement a risk identification method.

[0009] At least one embodiment of this specification can achieve the following beneficial effects: By acquiring image data to be identified including three-dimensional image data and two-dimensional image data, and performing image conversion processing on the three-dimensional image data to obtain converted two-dimensional image data; then, matching the converted two-dimensional image data with the two-dimensional image data in the image data to be identified, if the converted two-dimensional image data does not match the two-dimensional image data included in the image data to be identified, it can be determined that the image data to be identified has risks. Risk identification is performed in combination with three-dimensional image data and two-dimensional image data, thereby improving the accuracy of risk identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0011] Figure 1 This is a schematic diagram of an application scenario of a risk identification method provided in an embodiment of this specification; Figure 2 is a flow chart of a risk identification method provided by an embodiment of this specification; Figure 3 It is a swim lane diagram of a risk identification method provided by an embodiment of this specification; Figure 4 It is a schematic diagram of the structure of a risk identification device provided in an embodiment of this specification; Figure 5 It is a structural schematic diagram of a risk identification device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0012] In order to make the purpose, technical solutions and advantages of one or more embodiments of this specification clearer, the technical solutions of one or more embodiments of this specification will be clearly and completely described below in combination with the specific embodiments of this specification and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of one or more embodiments of this specification.

[0013] To facilitate understanding of the embodiments of this specification, the following is an explanation of the terms involved in the embodiments of this specification.

[0014] Face recognition: refers to the technology of identifying individual identity by detecting and analyzing facial features.

[0015] Anti-injection: refers to the system's ability to resist attack behaviors, such as high-definition screen and image injection attacks.

[0016] Depth information: refers to the Z-axis information of an object captured by a 3D camera, which is used to distinguish between real objects and two-dimensional images or videos. For example, a 3D camera is used to capture a face image and obtain the Z-axis information of the face image, which can be used to distinguish between a real face and a two-dimensional face image or video.

[0017] Two-dimensional image: A flat image that can be captured using an ordinary camera. A two-dimensional image captures a flat perspective of a scene, usually with only two dimensions of width and height.

[0018] Three-dimensional image: contains not only width and height information, but also depth information, which can more realistically represent the spatial position and shape of an object. In practical applications, the distance and shape of an object can be measured through structured light, time of flight (ToF), binocular vision and other technologies, and then a 3D model can be constructed to obtain a three-dimensional image. In face recognition, a three-dimensional face image can identify the user's facial contours, bumps and subtle facial features, etc., thereby improving the accuracy and security of facial recognition.

[0019] The technical solutions provided by the embodiments of this specification are described in detail below in conjunction with the accompanying drawings.

[0020] At present, face recognition and other image recognition technologies are commonly used in the fields of mobile payment and identity authentication. When performing face recognition and other image recognition, a terminal is usually used to collect a user's face image and upload the face image to a server so that the server can identify the acquired face image. Since the terminal may be subject to injection attacks by attackers during the process of uploading the face image to the server, it may cause property loss or leakage of sensitive information of the user. For example, an attacker may illegally obtain user resources or information through image injection attacks.

[0021] In order to solve the defects in the prior art, this solution provides the following embodiments.

[0022] Figure 1 It is a schematic diagram of an application scenario of a risk identification method provided in an embodiment of this specification.

[0023] like Figure 1 As shown, the solution may include a terminal 101 and a server 102; the terminal 101 may be a wearable terminal such as a mobile phone, a computer, or a smart watch; it may also be an access control terminal such as an access control or a gate for restricting access; it may also be a payment terminal such as a self-service vending machine, a self-service checkout device, or a POS machine for payment. The server 102 may be a server for an application carried by the terminal 101, or it may also be a system server for the terminal 101. The server 102 may be an independent physical server, or a server cluster or a distributed file system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0024] In practical applications, the terminal 101 may be provided with an image acquisition module to acquire image data; after acquiring image data using the image acquisition module, the terminal 101 may upload the acquired image data to the server. The server 102 has an image recognition function and may perform risk recognition on the acquired image data.

[0025] Next, a risk identification method provided in an embodiment of the specification will be described in detail with reference to the accompanying drawings.

[0026] Figure 2 This is a flow chart of a risk identification method provided by an embodiment of this specification. From a program perspective, the execution subject of the process can be a program installed on an application server or terminal.

[0027] like Figure 2 As shown, the process may include the following steps: Step 202: Obtain image data to be identified provided by the terminal.

[0028] In the embodiments of this specification, the terminal may be a device terminal capable of performing business processing; for example, a mobile phone, a computer, a smart watch, a cash register, a self-service checkout device, a door access control, a gate, etc. In addition, the terminal may include an image acquisition module so that the terminal can acquire image data to be identified. Specifically, the image acquisition module may be provided in the terminal; it may also be provided separately from the terminal, but it may be associated with the terminal, which is not specifically limited here.

[0029] In actual applications, under normal circumstances, the image data to be identified can represent the image data collected by the terminal and sent to the server. If the terminal is illegally attacked, the image data to be identified can also represent the image data provided by the attacker, which can be called injection attack image data or injection attack video stream. That is, if the terminal is illegally attacked, the image data to be identified is not the image or video normally collected by the terminal. For example, the attacker can hijack the image acquisition module, block the real image data of the image acquisition module from being sent to the server, and send false image data, that is, injection attack image data, to the server as the image data to be identified.

[0030] In actual applications, the terminal can collect multiple frames of image data; to improve recognition accuracy, the terminal or server can select at least one frame of image data with better quality from the multiple frames of image data as the image data to be identified. For example, after the terminal collects multiple frames of image data, it can use the image quality detection model to perform clarity detection, subject recognition, composition analysis, exposure analysis and other processing on the multiple frames of image data, so as to filter out one or more frames of image data with better image quality from the multiple frames of image data as the image data to be identified, and send the filtered image data to be identified to the server. In addition, after the terminal collects multiple frames of image data, it can also directly upload multiple frames of image data to be identified to the server. Then, the server uses the image quality model to filter out one or more frames of image data with better image quality from the multiple frames of image data as the image data to be identified, and perform risk identification. In the embodiments of this specification, the image quality model can be a model set in the server, or it can be called by the server, which is not specifically limited here.

[0031] The image data to be identified obtained by screening using the image quality inspection model can be one frame of image or multiple frames of image. If it is multiple frames of image, risk identification can be performed for each frame of image, and then the final risk identification result can be determined based on the risk identification result of each frame of image. For example, the final risk identification result can be determined based on the proportion of risky images in multiple frames of image; if the proportion of risky images is greater than or equal to a preset proportion, it can be determined that the image data to be identified has risks; if the proportion of risky images is less than a preset proportion, it can be determined that the image data to be identified does not have risks.

[0032] Step 204: performing image conversion processing on the three-dimensional image data included in the image data to be identified to obtain converted two-dimensional image data.

[0033] In the embodiments of this specification, the terminal may be a terminal with a three-dimensional image acquisition function. The image data to be identified may include three-dimensional image data; the three-dimensional image data may be image data containing depth information, and specifically, the three-dimensional image data may be RGB-D data or point cloud data. RGB-D data may include color data and image depth data, wherein the color data may include values ​​of three colors: red, green, and blue.

[0034] The two-dimensional image data may be planar image data containing image features of two dimensions, but not containing depth information. The converted two-dimensional image data may represent planar image data obtained by two-dimensionally converting the three-dimensional image data contained in the image data to be recognized.

[0035] Step 206: Determine whether the converted two-dimensional image data matches the two-dimensional image data included in the image data to be identified.

[0036] In the embodiments of this specification, the image data to be identified may also include two-dimensional image data. In practical applications, under normal circumstances, the two-dimensional image data may be collected by the terminal. In the case of an attack, the two-dimensional image data may be image data injected by the attacker.

[0037] Considering that in actual applications, attackers usually perform illegal operations by injecting two-dimensional images, rather than forging three-dimensional image data, the three-dimensional image data in the embodiments of this specification may be three-dimensional image data actually collected by the terminal.

[0038] In the embodiments of the present specification, a computer model can be used to determine whether the converted two-dimensional image data matches the two-dimensional image data in the image data to be identified. Specifically, the computer model can be a similarity matching model.

[0039] Step 208: If the converted two-dimensional image data does not match the two-dimensional image data included in the image data to be identified, it is determined that the image data to be identified has a risk.

[0040] If the converted two-dimensional image data does not match the two-dimensional image data contained in the image data to be identified, it can be indicated that the three-dimensional image data contained in the image data to be identified and the two-dimensional image data contained in the image data to be identified are image data of different objects. For example, the three-dimensional image data is the facial image data of user A, and the two-dimensional image data is the facial image data of user B. In this way, it can be determined that the image data to be identified are not image data of the same object, and therefore, it can be determined that the image data to be identified is at risk. In practical applications, when the image data to be identified is at risk of injection attack, it will cause the three-dimensional image data to represent a different object than the two-dimensional image data. Therefore, the risk of the image data to be identified may be an injection image attack risk.

[0041] Injection attack refers to the attacker injecting attack code into the place where the application or server receives terminal data in order to perform arbitrary operations, tamper with data or obtain sensitive information. In the embodiments of this specification, injection image attack refers to the attacker injecting attack image code into the place where the application or server receives terminal data.

[0042] In actual applications, for the three-dimensional image data and the two-dimensional image data included in the image data to be identified, the terminal may send the three-dimensional image data and the two-dimensional image data to the server together, or send the three-dimensional image data and the two-dimensional image data to the server separately. The three-dimensional image data and the two-dimensional image data may include the terminal identification of the terminal or the user identification of the user, so that the server can determine the three-dimensional image data and the two-dimensional image data that need to be identified.

[0043] Since the three-dimensional image data and the two-dimensional image data in the image data to be identified are image data collected for the same object, and the converted two-dimensional image data is obtained by image conversion processing based on the three-dimensional image data, therefore, if the two-dimensional image data in the image data to be identified obtained by the server has not been attacked, the converted two-dimensional image data should match the two-dimensional image data in the image data to be identified. If the two-dimensional image data in the image data to be identified obtained by the server has been attacked, for example, the attacker replaces or tampers with the two-dimensional image data in the image data to be identified with other two-dimensional image data, the converted two-dimensional image data does not match the two-dimensional image data in the image data to be identified. Therefore, if the converted two-dimensional image data does not match the two-dimensional image data in the image data to be identified, it can be determined that there is a risk in the image data to be identified.

[0044] It should be understood that the order of some steps in the methods described in one or more embodiments of this specification can be interchanged according to actual needs, or some steps can be omitted or deleted.

[0045] Figure 2 The method in the invention obtains image data to be identified including three-dimensional image data and two-dimensional image data, and performs image conversion processing on the three-dimensional image data to obtain converted two-dimensional image data; then, the converted two-dimensional image data is matched with the two-dimensional image data in the image data to be identified, so as to determine that the image data to be identified has risks when the converted two-dimensional image data does not match the two-dimensional image data included in the image data to be identified. The three-dimensional image data and the two-dimensional image data are combined to identify whether the image data to be identified has risks, thereby improving the accuracy of risk identification.

[0046] based on Figure 2 The method, the examples of this specification also provide some specific implementation plans of the method, which are described below.

[0047] Optionally, after obtaining the image data to be identified, the following steps may also be included: It is determined whether the image data to be identified includes image depth data.

[0048] If the image data to be identified does not include image depth data, it is determined that the image data to be identified has risks.

[0049] The depth information may be information of the Z axis in the three-dimensional image data. Specifically, the image depth data may represent distance information between different parts of the acquired object, and this information may be used to extract the three-dimensional shape of the acquired object.

[0050] For example, for a face image, the depth information in a 3D face image can be understood as the distance information between different parts of the face. This information can be used to extract the three-dimensional shape of the face, including the height, concavity, contours and distances between contours.

[0051] If the image data to be identified does not contain image depth data, it may mean that the image data to be identified does not include three-dimensional image data, and thus it can be determined that the image data to be identified is not the image actually captured by the terminal, the server has not received the image actually captured by the terminal, and there is a risk in the image data to be identified.

[0052] Correspondingly, performing image conversion processing on the three-dimensional image data contained in the image data to be identified to obtain converted two-dimensional image data may specifically include: If the image data to be identified includes image depth data, three-dimensional image data including the image depth data is acquired.

[0053] The three-dimensional image data is subjected to image conversion processing to obtain converted two-dimensional image data.

[0054] In practical applications, it is possible to determine whether the image data to be identified contains image depth data by determining whether the image data to be identified contains depth feature identifiers. It is also possible to determine whether the image data to be identified contains image depth data by determining whether the image information dimension contained in the image data to be identified has a depth dimension; for example, it is determined whether the image information dimension contained in the image data to be identified is three dimensions, and the values ​​of the data in any dimension are not all 0; if so, it can be determined that the image data to be identified contains image depth data; if not, it can be determined that the image data to be identified does not contain image depth data.

[0055] If the image data to be identified does not include image depth data, it can be determined that the image data to be identified does not include three-dimensional image data, and thus it can be determined that the image data to be identified has risks, and there is no need to perform the above steps 204 and 206.

[0056] In the embodiment of this specification, after the image data to be identified is obtained, it is first determined whether the image data to be identified contains the image depth data that should be included, so as to determine whether the image data to be identified contains three-dimensional image data, thereby preliminarily determining whether the image data to be identified has risks. Then, when the image data to be identified contains three-dimensional image data, image conversion processing is performed on the three-dimensional image data to obtain converted two-dimensional image data, and the converted two-dimensional image data is matched with the two-dimensional image data contained in the image data to be identified, so as to further determine whether the image data to be identified has risks based on the matching results. The embodiment of this specification constructs a multi-level risk identification mechanism, which further improves the security of risk identification.

[0057] For ease of understanding, the embodiments of this specification also provide a specific method for converting three-dimensional image data into two-dimensional image data.

[0058] Optionally, performing image conversion processing on the three-dimensional image data included in the image data to be identified to obtain converted two-dimensional image data may specifically include: The three-dimensional image data contained in the image data to be identified is projected and transformed to obtain converted two-dimensional image data.

[0059] In the embodiments of this specification, an orthographic projection conversion method or a perspective projection conversion method can be used to perform projection transformation processing on the three-dimensional image data to obtain converted two-dimensional image data. In practical applications, other methods or methods can also be used to perform projection transformation processing on the three-dimensional image data. Of course, other methods for converting three-dimensional image data into two-dimensional image data can also be used to convert two-dimensional image data, which is not specifically limited here.

[0060] In the embodiment of the present specification, the above method is to process the three-dimensional image data to obtain converted two-dimensional image data, and use the converted two-dimensional image data to match the two-dimensional image data in the image data to be identified, so as to determine whether the image data to be identified has risks.

[0061] In practical applications, the two-dimensional image data contained in the image data to be identified can also be converted into three-dimensional image data, and then the three-dimensional image data obtained by the conversion can be compared with the three-dimensional image data contained in the image data to be identified to determine whether there is a risk.

[0062] As an embodiment, the image data to be identified may include two-dimensional image data of a target object at multiple different angles. Then, the two-dimensional image data at multiple different angles are spliced ​​to obtain simulated image depth data of the target object. Then, it can be determined whether the image data to be identified is at risk based on whether the simulated image depth data matches the image depth data in the image data to be identified. If the simulated image depth data matches the image depth data in the image data to be identified, it can be determined that there is no risk in the image data to be identified; if the simulated image depth data does not match the image depth data in the image data to be identified, it can be determined that there is a risk in the image data to be identified.

[0063] As another embodiment, the simulated image depth data may be combined with the two-dimensional image data to obtain simulated three-dimensional image data. Whether the image data to be identified has a risk may be determined based on whether the simulated three-dimensional image data matches the three-dimensional image data in the image data to be identified. If the simulated three-dimensional image data matches the three-dimensional image data in the image data to be identified, it may be determined that the image data to be identified has no risk; if the simulated three-dimensional image data does not match the three-dimensional image data in the image data to be identified, it may be determined that the image data to be identified has a risk.

[0064] In the embodiments of this specification, specific contents on how to judge whether the converted two-dimensional image data matches the two-dimensional image data included in the image data to be identified are also provided.

[0065] Optionally, the determining whether the converted two-dimensional image data matches the two-dimensional image data included in the image data to be identified may specifically include: The similarity between the converted two-dimensional image data and the two-dimensional image data is calculated using a similarity calculation model.

[0066] It is determined whether the similarity is less than or equal to a preset threshold.

[0067] In the embodiment of the present specification, if the similarity between the converted two-dimensional image data and the two-dimensional image data in the image data to be identified is less than or equal to the preset similarity value, it can be determined that the converted two-dimensional image data does not match the two-dimensional image data, and further, it can be determined that the data to be identified has a risk. If the similarity between the converted two-dimensional image data and the two-dimensional image data is greater than the preset similarity value, it can be determined that the converted two-dimensional image data matches the two-dimensional image data, and further, it can be determined that the data to be identified has no risk. The preset similarity value can be set according to actual needs and is not specifically limited here.

[0068] In the embodiments of this specification, the similarity calculation model may be a computer model that can calculate the similarity of image data. Specifically, the similarity calculation model may be a structural similarity model, an image hash algorithm model, or a pre-trained neural network model, such as VGG16, ResNet, etc.

[0069] In practical applications, the similarity of images can also be determined by comparing feature points. Specifically, for the converted two-dimensional image data and the two-dimensional image data contained in the image data to be identified, the area where the first target object is located in the converted two-dimensional image data can be first detected to obtain a first image containing the first target object; the area where the first target object is located in the two-dimensional image data can be detected to obtain a second image containing the first target object; then, the first image and the second image are aligned, and feature points are extracted from the aligned first image and the second image; the similarity of the first image and the second image is determined by the feature points, and the similarity can represent the similarity between the converted two-dimensional image data and the two-dimensional image data. The first target object can be a human biometric feature. Human biometric features can include face, fingerprint, palm print, iris, etc.

[0070] In practical applications, the alignment processing for the first image and the second image may include cropping, rotating, and other operations on the first image or the second image so that the first image and the second image have the same size and angle. Specifically, the first image and the second image may be aligned with a certain feature point as the center. When the image data to be identified is face image data, the first image and the second image may be images of the area where the face is located. At this time, the first image and the second image may be aligned with the feature point of the center of the nose tip as the center. Of course, the first image and the second image may also be aligned with other feature points as the center point, which is not specifically limited here.

[0071] In an embodiment of the present specification, the similarity between the converted two-dimensional image data and the two-dimensional image data is calculated based on the feature points in the first image and the feature points in the second image, which may specifically include: calculating the similarity coefficient between each feature point in the first image and each feature in the second image; determining the number of feature points whose similarity coefficient is greater than a preset similarity coefficient; determining the similarity between the converted two-dimensional image data and the two-dimensional image data based on the ratio of the number of feature points whose similarity coefficient is greater than the preset similarity coefficient to the total number of feature points. If the similarity between the converted two-dimensional image data and the two-dimensional image data is less than or equal to a preset threshold, it can be determined that the converted two-dimensional image data does not match the two-dimensional image data, and further, it can be determined that the image data to be identified has a risk. If the similarity between the converted two-dimensional image data and the two-dimensional image data is greater than the preset threshold, it can be determined that the converted two-dimensional image data matches the two-dimensional image data, and further, it can be determined that the image data to be identified has no risk.

[0072] The specific points can be set according to actual needs. If the first image and the second image are face images, the feature points may include center points or edge points of eyes, nose, mouth and other parts. In the embodiment of this specification, in order to improve the accuracy of determining whether the converted two-dimensional image data matches the two-dimensional image data, a preset number of feature points may be selected to perform similarity calculation; for example, more than 100 feature points may be selected.

[0073] At present, in the process of image recognition, in order to ensure that the object of image recognition is a real person rather than a photo or other form of image, in order to ensure the accuracy of risk identification, liveness recognition is usually required. At present, when performing liveness recognition, it is usually necessary to interact with the user and make the user cooperate to perform corresponding actions; for example, blinking, opening the mouth, turning the head to the left, turning the head to the right, etc., the process is cumbersome. In the embodiment of this specification, the method of converting three-dimensional image data into two-dimensional image data and comparing it with the two-dimensional image data in the image data to be identified can also be used as a liveness recognition method. And there is no need for the user to perform too many actions, which is simpler and more efficient.

[0074] The image data to be identified in the embodiments of this specification may be an image including biometric features, and the specific biometric features may be human biometric features with depth or specific patterns, such as face, iris, fingerprint, palm print, etc. The distribution of facial features may also represent the distribution of features with depth, and palm prints, fingerprints, etc. may represent features with specific patterns.

[0075] As an implementation manner, the image data to be recognized includes face image data, and the method may further include: It is determined whether the three-dimensional image data contained in the image data to be recognized conforms to the face distribution characteristics.

[0076] If the three-dimensional image data does not conform to the facial distribution characteristics, it is determined that the image data to be identified is risky.

[0077] In the embodiments of the present specification, if the biometric feature is a face, before performing image conversion processing on the three-dimensional image data, it is also possible to determine whether the three-dimensional image data contained in the image data to be identified conforms to the face distribution characteristics; it is understandable that determining whether the three-dimensional image data contained in the image data to be identified conforms to the face distribution characteristics may also be performed at the same time as or after performing image conversion processing on the three-dimensional image data, and is not specifically limited here.

[0078] In practical applications, determining whether the three-dimensional image data included in the image data to be identified conforms to the face distribution characteristics may specifically be: determining whether the image depth data in the image data to be identified conforms to the face distribution characteristics.

[0079] If the biometric feature is a face, the image data to be identified may include face image data; specifically, the image data to be identified may include three-dimensional face image data and two-dimensional face image data. Since the facial features are not on the same plane, the image depth data of different pixels in the three-dimensional face image data collected by the terminal should be different, and the image depth data should conform to the distribution characteristics of the facial features; for example, from the overall point of view, the depth value at the nose position of the face should be smaller than the depth value at the eye position, and the depth value at the mouth position should be smaller than the depth value at the eye position. From the details, the depth value at the inner corner of the eye of the face should be smaller than the depth value at the outer corner; the depth value at the edge of the mouth should be smaller than the depth value at the center of the mouth. Therefore, the real face detection can be performed based on the image depth data in the image data to be identified to preliminarily determine whether there is a risk in the image data to be identified.

[0080] In the embodiments of this specification, if the three-dimensional image data conforms to the face distribution, it can be preliminarily determined that the three-dimensional image data is a real face, and then it can be preliminarily determined that the image data to be identified does not have risks; if the three-dimensional image data does not conform to the face distribution, it can be determined that the three-dimensional image data is not a real face, but is false, and then it can be determined that the image data to be identified has risks. The method of judging whether the object to be image recognized is a real face based on whether the three-dimensional image data conforms to the face distribution, and then determining whether the image data to be identified has risks, avoids the process of interacting with the user and asking the user to perform related actions, simplifies the steps of face detection and risk identification of the image data to be identified, and improves the user experience.

[0081] In addition, whether the image depth data in the three-dimensional image data conforms to the facial distribution characteristics can be used to identify whether the object of image recognition is a real person. This also reduces the impact of image color on detection, that is, reduces the impact of lighting conditions on detection, so that detection can be performed in various lighting environments. At the same time, it improves the accuracy of detection in environments with poor lighting, thereby improving the accuracy of risk identification of image data to be identified.

[0082] If the 3D image data conforms to the distribution of human faces, it can be preliminarily determined that the 3D image data is authentic and can be used to match with the 2D image data.

[0083] Optionally, performing image conversion processing on the three-dimensional image data included in the image data to be identified may specifically include: If the three-dimensional image data conforms to the face distribution characteristics, image conversion processing is performed on the three-dimensional image data contained in the image data to be identified.

[0084] In practical applications, if it is determined that the image data to be identified has risks, a prompt message may be generated and sent to the terminal so that the user is aware of the current risks. Optionally, after determining that the image data to be identified has risks, the following may also be included: Generates a prompt message indicating that there is a risk or that the business verification fails.

[0085] Send the prompt information to the terminal.

[0086] In the embodiments of this specification, the prompt information may be at least one of visual prompt information, auditory prompt information, and tactile prompt information. Specifically, the visual prompt information may be at least one of text prompt information and animation prompt information; the auditory prompt information may be voice prompt information; and the tactile prompt information may be vibration prompt information.

[0087] In actual applications, when using image data to unlock a terminal, the prompt information may be information used to prompt the user that the image recognition fails and the device is not unlocked. For example, when using a face image or fingerprint image to unlock a terminal, the prompt information may be “There is a risk in face verification and the device cannot be unlocked. Please try again” or “There is a risk in fingerprint verification and the device cannot be unlocked. Please try again”.

[0088] In the process of payment using image data, the prompt information may be information used to prompt the user that the image recognition has failed and the payment has failed; for example, in the process of payment using a face image or fingerprint image, the prompt information may specifically be "There is a risk in face verification, payment has failed, please try again" or "There is a risk in fingerprint verification, payment has failed, please try again", etc.

[0089] In the process of authorization using image data, the prompt information may be information used to prompt the user that the image recognition has failed and the authorization has failed; for example, in the process of authorization using a face image or a fingerprint image, the prompt information may specifically be "There is a risk in face verification, authorization has failed, please try again" or "There is a risk in fingerprint verification, authorization has failed, please try again", etc.

[0090] The above-mentioned specific prompt information, such as "There is a risk in face verification and cannot be unlocked, please try again" or "There is a risk in fingerprint verification and cannot be unlocked, please try again", can be at least one of text form and voice form, and a vibration prompt information can also be output at the same time, before or after outputting the above-mentioned specific prompt information to prompt the user.

[0091] In actual applications, after determining that there is a risk in the image data to be identified and sending a prompt message to the terminal, the business process related to image recognition can be ended according to the user's choice; in order to avoid user losses, the business process related to image recognition can also be automatically ended without the need for user selection.

[0092] Optionally, the method may further include: If the converted two-dimensional image data matches the two-dimensional image data included in the image data to be identified, it is determined that there is no risk in the image data to be identified.

[0093] The fact that the image data to be identified does not have risks can indicate that the image data to be identified is image data actually collected by the terminal, and there is no risk of the image being tampered with. If the image data to be identified does not have risks, the subsequent business processing flow can be carried out.

[0094] If it is determined that the converted two-dimensional image data matches the two-dimensional image data contained in the image data to be identified, it can only be determined that the three-dimensional image data and the two-dimensional image data are image data for the same object, which can also indicate that the three-dimensional image data and the two-dimensional image data are image data actually collected by the terminal, but it cannot be determined whether the image data to be identified is for the business user himself. For example, in actual applications, when a user handles a business or unlocks a terminal, further identity recognition is required to determine whether it is the account user himself or the terminal user himself.

[0095] Optionally, after determining that the image data to be identified does not have risks in the embodiments of this specification, the following steps may also be included: The three-dimensional image data contained in the image data to be identified and / or the two-dimensional image data contained in the image data to be identified are compared with the image data in the business image database to determine whether the image data to be identified is the image data of the target user.

[0096] In the embodiments of this specification, the business image database is a database for storing the actual user identification and the real image data of the user; the actual user identification and the user's image data can be stored in correspondence. The actual user identification can be a user account; specifically, the account and any one or more of the user's face image, fingerprint image, and palm print image can be stored in correspondence. In a business system for unlocking a terminal, the actual user identification can also be a terminal identification. In order to facilitate matching the image data to be identified with the image data in the database during image recognition, the feature points of the image data can also be stored in correspondence with the actual user identification. The business image database can be stored on a server or locally on a terminal.

[0097] In practical applications, the feature points of the two-dimensional image data or the three-dimensional image data in the image data to be identified can be first extracted; then the similarity between the feature points of the extracted two-dimensional image data or the three-dimensional image data and the feature points of the user's own image data stored in the database can be calculated; if the similarity exceeds a preset similarity threshold, it can be determined that the image data to be identified is the user's own image data.

[0098] In the embodiment of this specification, if the terminal cannot collect a three-dimensional image, but the image data to be identified obtained by the server contains three-dimensional image data, it can be determined that the three-dimensional image data contained in the image data to be identified was injected by an attacker through illegal means.

[0099] Based on this, the method may further include: It is determined whether the terminal has a function of acquiring three-dimensional images.

[0100] The performing image conversion processing on the three-dimensional image data included in the image data to be identified may specifically include: If the terminal has a function of acquiring three-dimensional images, image conversion processing is performed on the three-dimensional image data included in the image data to be identified.

[0101] In the embodiments of this specification, it can be determined whether the terminal has a three-dimensional image acquisition function according to the terminal identification of the terminal. The terminal identification can be at least one of a terminal system identification and a terminal hardware identification. The terminal system identification is identification information representing the terminal system, such as a system model, a system name, a system version number, etc. The terminal hardware identification can be a terminal brand, a terminal model, terminal provider information, etc.

[0102] Specifically, the terminal identification of the terminal with the 3D image acquisition function can be determined first, and then a 3D image acquisition function whitelist can be established for the terminal identification with the 3D image acquisition function. In practical applications, as the terminal products are updated, the terminal identification in the 3D image acquisition function whitelist can be updated regularly or irregularly according to the update status of the terminal. In addition, the 3D image acquisition function whitelist can be stored in the server for easy use by the server.

[0103] When the server obtains the image data to be identified from the terminal, it can obtain the terminal identification of the terminal. The server can determine whether the terminal has the 3D image acquisition function according to the obtained terminal identification and the 3D image acquisition function whitelist.

[0104] In the embodiment of the present specification, if the terminal does not have a three-dimensional image acquisition function, but the image data to be identified obtained by the server contains three-dimensional image data, it can be determined that the three-dimensional image data contained in the image data to be identified was injected by an attacker through illegal means, and thus it can be determined that there is a risk in the image data to be identified.

[0105] As another embodiment, if the terminal does not have a three-dimensional image acquisition function, and the image data to be identified obtained by the server does not contain three-dimensional image data, the two-dimensional image data contained in the image data to be identified can be directly compared with the image data in the business image database to determine whether the image data to be identified is the user's own image data; if it is determined that the image data to be identified is the user's own image data, it can be determined that there is no risk in image recognition, and the business process related to image recognition can continue to be executed. For example, if the image data to be identified is obtained during the payment request process, and it is determined that there is no risk in the image data to be identified, and at the same time, the image data to be identified is also the user's own image data, at this time, the resources used for payment can be transferred to the resource recipient.

[0106] Figure 3 This is a swim lane diagram of a risk identification method provided in an embodiment of this specification. Taking face recognition and face image data risk detection as an example, it is explained. Figure 3 As shown, the risk identification process may involve execution entities such as terminals and servers, and the process may include a phase of acquiring facial image data to be identified and a phase of risk identification of facial image data to be identified.

[0107] In the phase of acquiring the face image data to be identified, the execution subject may include a terminal, and specifically may include the following steps: Step 302: Obtaining face image data to be recognized; The terminal may be a terminal with a function of collecting three-dimensional images. The terminal may collect multiple frames of three-dimensional face image data and multiple frames of two-dimensional face image data; then, the image quality detection model may be used to perform quality detection on the multiple frames of three-dimensional face image data and the multiple frames of two-dimensional face image data, so as to screen out one frame of three-dimensional face image data and one frame of two-dimensional face image data with better image quality from the multiple frames of three-dimensional face image data and the multiple frames of two-dimensional face image data, respectively. The screened three-dimensional face image data and two-dimensional face image data are sent to the server as face image data to be identified.

[0108] In the risk identification stage of the face image data to be identified, the execution subject may include a server, and the server may be a server of an application. In the risk identification stage of the face image data to be identified, the following steps may be specifically included: Step 304: Obtain the facial image data to be recognized sent by the terminal.

[0109] In the embodiments of the present specification, under normal circumstances, the two-dimensional facial image data in the facial image data to be identified obtained by the server may be collected by the terminal; however, in the event of an attack, the two-dimensional facial image data in the facial image data to be identified obtained by the server may be image data injected by the attacker.

[0110] Step 306 : Determine whether the face image data to be recognized contains image depth data; if not, execute step 308 ; if so, execute step 310 .

[0111] The depth information of an image is the Z-axis information in the three-dimensional image data. Specifically, for a face image, if you imagine a face as being made up of many points, then each point in the three-dimensional data has an actual physical distance from the observer to the point, which is the "depth" of the point. The depth data of all these points together constitute the depth map or depth information of the entire face.

[0112] In the embodiments of this specification, the process of determining whether the image data of the face to be recognized contains image depth data can refer to the relevant content described above, and will not be repeated here.

[0113] Step 308: Determine whether the facial image data to be identified is risky.

[0114] If the facial image data to be identified does not contain image depth data, it may mean that the facial image to be identified is not the image actually captured by the terminal, and the server has not received the image actually captured by the terminal. It can be determined that there is a risk in the facial image data to be identified.

[0115] If it is determined that the facial image data to be identified is at risk, the server may also send risk warning information to the terminal.

[0116] Step 310: Determine whether the three-dimensional face image data contained in the face image data to be identified conforms to the face distribution characteristics; if not, execute step 308; if yes, execute step 312.

[0117] In the embodiment of the present specification, determining whether the three-dimensional face image data contained in the face image data to be identified conforms to the face distribution characteristics is similar to the method of determining whether the three-dimensional face image data contained in the image data to be identified conforms to the face distribution characteristics in the previous text, and will not be repeated here.

[0118] In the embodiments of the present specification, if the three-dimensional face image data does not conform to the face distribution, it can be preliminarily determined that the three-dimensional face image data is false, and thus it can be preliminarily determined that there is a risk in the face image data to be identified; if the three-dimensional face image data conforms to the face distribution, it can be preliminarily determined that the three-dimensional face image data is true, and thus it can be preliminarily determined that there is no risk in the face image to be identified.

[0119] Step 312: Perform image conversion processing on the three-dimensional face image data to obtain converted two-dimensional face image data.

[0120] In the embodiments of this specification, the specific method of performing image conversion processing on the three-dimensional face image data to obtain the converted two-dimensional face image data can be found in the relevant content in the previous text, and will not be repeated here.

[0121] Step 314: Determine whether the converted two-dimensional facial image data matches the two-dimensional facial image data included in the facial image data to be identified; if not, execute step 308; if yes, execute step 316.

[0122] In the embodiments of this specification, the specific method for determining whether the converted two-dimensional facial image data matches the two-dimensional facial image data included in the facial image data to be identified can be found in the relevant content in the previous text and will not be repeated here.

[0123] In the embodiments of the present specification, if the converted two-dimensional face image data does not match the two-dimensional face image data included in the face image data to be identified, it can be indicated that the three-dimensional face image data included in the face image data to be identified and the two-dimensional face image data included in the face image data to be identified are image data of different faces. For example, the three-dimensional face image data is the face image data of user A, and the two-dimensional face image data is the face image data of user B. In this way, it can be determined that the face image data to be identified are not image data of the same face, and therefore, it can be determined that the face image data to be identified is risky.

[0124] In an embodiment of the present specification, if the converted two-dimensional facial image data matches the two-dimensional facial image data contained in the facial image data to be identified, it can be indicated that the three-dimensional facial image data contained in the facial image data to be identified and the two-dimensional facial image data contained in the facial image data to be identified are image data of the same face, and it can be determined that there is no risk in the facial image data to be identified.

[0125] Step 316: Determine that there is no risk in the facial image data to be identified.

[0126] In actual applications, if it is determined that there is no risk in the facial image data to be identified, the server can also perform subsequent business processing procedures based on the acquired two-dimensional image data and / or three-dimensional image data.

[0127] Based on the same idea, the embodiments of this specification also provide a device corresponding to the above method. Figure 4 This is a schematic diagram of a risk identification device provided in an embodiment of this specification. Figure 4 As shown, the device may include: The acquisition module 402 is used to acquire the image data to be recognized provided by the terminal.

[0128] The conversion module 404 is used to perform image conversion processing on the three-dimensional image data included in the image data to be recognized to obtain converted two-dimensional image data.

[0129] The first judgment module 406 is used to judge whether the converted two-dimensional image data matches the two-dimensional image data included in the image data to be recognized.

[0130] The first determination module 408 is configured to determine that the image data to be identified has a risk if the converted two-dimensional image data does not match the two-dimensional image data included in the image data to be identified.

[0131] based on Figure 4 The present specification also provides some specific implementation schemes of the method, which are described below.

[0132] Optional, Figure 4 The device may further include: The second judgment module is used to judge whether the image data to be identified contains image depth data.

[0133] The second determination module is used to determine that the image data to be identified has risks if the image data to be identified does not include image depth data.

[0134] The conversion module 404 may be specifically used for: If the image data to be identified includes image depth data, three-dimensional image data including the image depth data is acquired.

[0135] The three-dimensional image data is subjected to image conversion processing to obtain converted two-dimensional image data.

[0136] Optionally, the conversion module 404 may be specifically used to: The three-dimensional image data contained in the image data to be identified is projected and transformed to obtain converted two-dimensional image data.

[0137] Optionally, the first determining module 406 may be specifically configured to: The similarity between the converted two-dimensional image data and the two-dimensional image data is calculated using a similarity calculation model.

[0138] It is determined whether the similarity is less than or equal to a preset threshold.

[0139] Optionally, the image data to be identified includes face image data; Figure 4 The device may further include: The third judgment module is used to judge whether the three-dimensional image data contained in the image data to be recognized conforms to the face distribution characteristics.

[0140] The third determination module is used to determine that there is a risk in the image data to be identified if the three-dimensional image data does not conform to the face distribution characteristics.

[0141] Optionally, the conversion module 404 may be specifically used to: If the three-dimensional image data conforms to the face distribution characteristics, image conversion processing is performed on the three-dimensional image data contained in the image data to be identified.

[0142] Optionally, the Figure 4 The device may further include: The prompt information generation module is used to generate prompt information indicating that there is a risk or that the business verification fails.

[0143] A sending module is used to send the prompt information to the terminal.

[0144] Optionally, the Figure 4 The device may further include: The fourth determination module is used to determine that there is no risk in the image data to be identified if the converted two-dimensional image data matches the two-dimensional image data included in the image data to be identified.

[0145] Optionally, the Figure 4 The device may further include: The comparison module is used to compare the three-dimensional image data contained in the image data to be identified and / or the two-dimensional image data contained in the image data to be identified with the image data in the business image database to determine whether the image data to be identified is the image data of the target user.

[0146] Optionally, the Figure 4 The device may further include: The fourth determination module is used to determine whether the terminal has the function of acquiring three-dimensional images.

[0147] The conversion module 404 may be specifically used for: If the terminal has a function of acquiring three-dimensional images, image conversion processing is performed on the three-dimensional image data included in the image data to be identified.

[0148] Optionally, the first determining module 408 may be specifically configured to: It is determined that the image data to be identified has a risk of injection attack.

[0149] Based on the same idea, the embodiments of this specification also provide a device corresponding to the above method.

[0150] Figure 5 This is a schematic diagram of the structure of a risk identification device provided in an embodiment of this specification. Figure 5 As shown, the device 500 may include: at least one processor 510; and, A memory 530 in communication with the at least one processor; wherein, The memory 530 stores instructions 520 executable by the at least one processor 510. The instructions are executed by the at least one processor 510 to enable the at least one processor 510 to: Obtain the image data to be recognized provided by the terminal.

[0151] The three-dimensional image data contained in the image data to be identified is subjected to image conversion processing to obtain converted two-dimensional image data.

[0152] It is determined whether the converted two-dimensional image data matches the two-dimensional image data included in the image data to be recognized.

[0153] If the converted two-dimensional image data does not match the two-dimensional image data included in the image data to be identified, it is determined that the image data to be identified has a risk.

[0154] Based on the same idea, the embodiment of this specification also provides a computer-readable medium corresponding to the above method. The computer-readable medium stores computer-readable instructions, which can be executed by a processor to implement the above risk identification method: Each embodiment in this specification is described in a progressive manner. The same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. Figure 3 As for the device shown, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0155] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards in the relevant regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0156] In the 1990s, it was very clear whether the improvement of a technology was hardware improvement (for example, improvement of the circuit structure of diodes, transistors, switches, etc.) or software improvement (improvement of the method flow). However, with the development of technology, many improvements of the method flow today can be regarded as direct improvements of the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that the improvement of a method flow cannot be implemented with hardware entity modules. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit whose logical function is determined by the user's programming of the device. Designers can "integrate" a digital system on a PLD by programming it themselves, without having to ask chip manufacturers to design and make dedicated integrated circuit chips. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when developing and writing programs, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL). There is not only one kind of HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also know that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages ​​and program it into the integrated circuit, and then it is easy to obtain the hardware circuit that implements the logic method flow.

[0157] The controller may be implemented in any suitable manner, for example, the controller may take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (e.g., software or firmware) executable by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include but are not limited to the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, and the memory controller may also be implemented as part of the control logic of the memory. It is also known to those skilled in the art that, in addition to implementing the controller in a purely computer-readable program code manner, the controller may be implemented in the form of a logic gate, a switch, an application-specific integrated circuit, a programmable logic controller, and an embedded microcontroller by logically programming the method steps. Therefore, such a controller may be considered as a hardware component, and the devices for implementing various functions included therein may also be considered as structures within the hardware component. Or even, the devices for implementing various functions may be considered as both software modules for implementing the method and structures within the hardware component.

[0158] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0159] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0160] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0161] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0162] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0163] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0164] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0165] Memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0166] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules 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 disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined in this article, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0167] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device 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, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0168] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0169] The present application may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0170] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for risk identification, comprising: Obtain image data to be identified provided by the terminal; Performing image conversion processing on the three-dimensional image data contained in the image data to be identified to obtain converted two-dimensional image data; Determining whether the converted two-dimensional image data matches the two-dimensional image data included in the image data to be identified; If the converted two-dimensional image data does not match the two-dimensional image data included in the image data to be identified, it is determined that the image data to be identified has a risk.

2. The method according to claim 1, after obtaining the image data to be identified, further comprising: Determining whether the image data to be identified includes image depth data; If the image data to be identified does not include image depth data, determining that the image data to be identified has a risk; The performing image conversion processing on the three-dimensional image data contained in the image to be recognized to obtain converted two-dimensional image data specifically includes: If the image data to be identified includes image depth data, acquiring three-dimensional image data including the image depth data; The three-dimensional image data is subjected to image conversion processing to obtain converted two-dimensional image data.

3. The method according to claim 1, wherein the step of performing image conversion processing on the three-dimensional image data contained in the image data to be identified to obtain converted two-dimensional image data specifically comprises: The three-dimensional image data contained in the image data to be identified is projected and transformed to obtain converted two-dimensional image data.

4. The method according to claim 1, wherein determining whether the converted two-dimensional image data matches the two-dimensional image data included in the image data to be identified specifically comprises: Calculating the similarity between the converted two-dimensional image data and the two-dimensional image data using a similarity calculation model; It is determined whether the similarity is less than or equal to a preset threshold.

5. The method according to claim 1, wherein the image data to be identified comprises face image data; The method further comprises: Determining whether the three-dimensional image data contained in the image data to be recognized conforms to the face distribution characteristics; If the three-dimensional image data does not conform to the facial distribution characteristics, it is determined that the image data to be identified is risky.

6. The method according to claim 5, wherein the step of performing image conversion processing on the three-dimensional image data contained in the image data to be identified comprises: If the three-dimensional image data conforms to the face distribution characteristics, image conversion processing is performed on the three-dimensional image data contained in the image data to be identified.

7. The method according to claim 1, after determining that the image data to be identified has a risk, further comprising: Generate a prompt message indicating that there is a risk or that the business verification fails; Send the prompt information to the terminal.

8. The method of claim 1, further comprising: If the converted two-dimensional image data matches the two-dimensional image data included in the image data to be identified, it is determined that there is no risk in the image data to be identified.

9. The method according to claim 8, after determining that the image data to be identified does not have risks, further comprising: The three-dimensional image data contained in the image data to be identified and / or the two-dimensional image data contained in the image data to be identified are compared with the image data in the business image database to determine whether the image data to be identified is the image data of the target user.

10. The method of claim 1, further comprising: Determining whether the terminal has a function of acquiring three-dimensional images; The performing image conversion processing on the three-dimensional image data contained in the image data to be identified specifically includes: If the terminal has a function of acquiring three-dimensional images, image conversion processing is performed on the three-dimensional image data included in the image data to be identified.

11. The method according to claim 1, wherein determining that the image data to be identified has a risk comprises: It is determined that the image data to be identified has a risk of injection attack.

12. A risk identification device, comprising: An acquisition module, used for acquiring image data to be recognized provided by a terminal; A conversion module, used for performing image conversion processing on the three-dimensional image data contained in the image data to be identified to obtain converted two-dimensional image data; A first judgment module, used to judge whether the converted two-dimensional image data matches the two-dimensional image data included in the image data to be recognized; The first determination module is configured to determine that the image data to be identified has a risk if the converted two-dimensional image data does not match the two-dimensional image data included in the image data to be identified.

13. A risk identification device, comprising: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: Obtain image data to be identified provided by the terminal; Performing image conversion processing on the three-dimensional image data contained in the image data to be identified to obtain converted two-dimensional image data; Determining whether the converted two-dimensional image data matches the two-dimensional image data included in the image data to be identified; If the converted two-dimensional image data does not match the two-dimensional image data included in the image data to be identified, it is determined that the image data to be identified has a risk.

14. A computer-readable medium having computer-readable instructions stored thereon, wherein the computer-readable instructions can be executed by a processor to implement a risk identification method according to any one of claims 1 to 11.

Images