Identity verification method and device, storage medium and program product

Abstract

The invention discloses an identity verification method and device, a storage medium and a program product, and relates to the technical field of finance science and technology, and discloses an identity verification method, and the identity verification method comprises the steps: obtaining an original lossless image formed by the shooting of a camera in response to a human face verification instruction; and performing face verification processing based on the original lossless image to obtain an identity verification result. The technical problem that the financial security is reduced due to forged information generated by the existing AIGC technology is solved.

Description

Technical Field

[0001] The present application relates to the field of financial technology (Fintech), and in particular to an identity verification method, device, storage medium and program product. Background Art

[0002] In recent years, with the rapid advancement of artificial intelligence (AI) technology, especially the rise of AIGC (Artificial Intelligence Generated Content) technology, the protection of network security and information authenticity faces unprecedented challenges.

[0003] In the financial sector, this trend is particularly severe. As an information-intensive and high-risk industry, the financial industry has extremely high requirements for the authenticity of information and the security of user identities. However, the abuse of AIGC technology has opened up new avenues for financial fraud activities, such as illegal loans, credit card fraud, and insurance fraud through forged identity information, which seriously damages the interests of financial institutions and customers, and also poses a threat to the stability of the financial market. As a result, the forged information generated by existing AIGC technology reduces financial security.

[0004] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of this application is to provide an identity verification method, device, storage medium and program product, aiming to solve the technical problem that the forged information generated by the existing AIGC technology reduces financial security.

[0006] To achieve the above purpose, the present application proposes an identity verification method, which includes:

[0007] In response to a face verification instruction, an original lossless image captured by a camera is obtained;

[0008] Face verification processing is performed based on the original lossless image to obtain an identity verification result.

[0009] In some embodiments, the step of performing face verification processing based on the original lossless image to obtain an identity verification result includes:

[0010] Performing predetermined image processing on the original lossless image to obtain a preprocessed image;

[0011] Performing liveness detection on the preprocessed image to obtain a liveness detection result;

[0012] After the liveness detection result is that the liveness detection is passed, face comparison is performed on the pre-processed image to obtain an identity verification result.

[0013] In some embodiments, the step of performing liveness detection on the original lossless image to obtain a liveness detection result includes:

[0014] Inputting the preprocessed image into a preset living body detection model to obtain credibility;

[0015] After the liveness credibility is greater than a preset credibility threshold, the liveness detection passing is taken as the liveness detection result.

[0016] In some embodiments, before the step of inputting the preprocessed image into a preset living body detection model to obtain the credibility, the method includes:

[0017] Acquire a first training sample, wherein the first training sample comprises a first image sample in an original lossless format;

[0018] The first initial model is trained based on the first image sample until the first initial model converges to obtain a preset living body detection model.

[0019] In some embodiments, the step of performing face comparison on the original lossless image to obtain an identity verification result includes:

[0020] Obtain target facial information of the target person;

[0021] Inputting the preprocessed image and the target facial information into a preset face comparison model to obtain a matching degree between the preprocessed image and the target facial information;

[0022] After the matching degree is greater than a preset matching threshold, the identity verification is passed as the identity verification result.

[0023] In some embodiments, the step of inputting the preprocessed image and the target facial information into a preset face comparison model to obtain a matching degree between the preprocessed image and the target facial information includes:

[0024] Acquire a second training sample, wherein the second training sample comprises a second image sample in an original lossless format;

[0025] The second initial model is trained based on the second image sample until the second initial model converges to obtain a preset face comparison model.

[0026] In some embodiments, the step of performing predetermined image processing on the original lossless image to obtain a preprocessed image includes:

[0027] Obtaining a color filter array of the camera;

[0028] According to the arrangement positions of the primary colors in the color filter array, the original lossless image is split to obtain the primary color channel images as preprocessed images.

[0029] In addition, to achieve the above-mentioned purpose, the present application also proposes an identity verification device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the identity verification method described above.

[0030] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the identity verification method described above are implemented.

[0031] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the identity verification method described above.

[0032] One or more technical solutions proposed in this application have at least the following technical effects:

[0033] This application obtains the original lossless image formed by the camera shooting by responding to the face verification instruction, and then performs face verification processing based on the original lossless image to obtain the identity verification result. In this application, the original lossless image is used as the input of face verification. On the one hand, since the original lossless image is the original image formed by the camera capturing photoelectrons through a photosensitive element (such as CCD, CMOS, etc.), rather than the full-color image output after image signal processing of the original lossless image formed by the camera in the prior art, the original lossless image is encapsulated inside the camera, and it is difficult for attackers to attack the hardware bottom layer to tamper with the original lossless image. On the other hand, the current AIGC model is basically obtained by training with full-color images. Therefore, if the AIGC model is used to forge the original lossless image, it is necessary to invest a lot of technical costs and time costs to completely retrain the new model. In addition, since the original lossless image is the image data of the hardware bottom layer, the number of original lossless images on the market is small, and the training of the AIGC model requires an extremely large number of training samples to be possible. Therefore, in the case of insufficient training samples, the trained new model may also be completely unavailable, and the investment cost is much higher than the counterfeiting income. In addition, for the existing face verification process, this application does not make any other changes to the main process of the existing face verification process except replacing the input image from a full-color image to an original lossless image, so it has good compatibility with the existing face verification process. Therefore, this application performs face verification based on the original lossless image, which can effectively resist the attack of forged information generated by the existing AIGC technology and improve financial security. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0035] In order to more clearly illustrate the embodiments of the present application 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, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0036] Figure 1 The figure is a schematic diagram of the existing facial verification process;

[0037] Figure 2 A schematic diagram of a scenario for generating a full-color image based on an original lossless image;

[0038] Figure 3 A flowchart of the first embodiment of the identity verification method for this application is provided;

[0039] Figure 4 Schematic diagram of the array structure of the Bayer array involved in the embodiment of the present application;

[0040] Figure 5 A graphic showing the difference between the original lossless image and the full-color image;

[0041] Figure 6 A split scene graph of the original lossless image involved in the embodiment of the present application;

[0042] Figure 7 A flowchart of the second embodiment of the identity verification method for this application is provided;

[0043] Figure 8 A flowchart of the third embodiment of the identity verification method for this application is provided;

[0044] Fig. 9 A comparison diagram of the face verification scenes involved in the embodiments of the present application;

[0045] Fig.10 Schematic diagram of the device structure of the hardware operating environment involved in the identity verification method in the embodiment of the present application.

[0046] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0047] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0048] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0049] The main solution of the embodiment of the present application is: in response to a face verification instruction, an original lossless image captured by a camera is obtained; based on the original lossless image, face verification processing is performed to obtain an identity verification result.

[0050] In this embodiment, for ease of description, the following description is made with the identity verification device as the execution entity.

[0051] In recent years, with the rapid advancement of artificial intelligence (AI) technology, especially the rise of AIGC (Artificial Intelligence Generated Content) technology, the protection of network security and information authenticity faces unprecedented challenges.

[0052] In the financial sector, this trend is particularly severe. As an information-intensive and high-risk industry, the financial industry has extremely high requirements for the authenticity of information and the security of user identities. However, the abuse of AIGC technology has opened up new avenues for financial fraud activities, such as illegal loans, credit card fraud, and insurance fraud through forged identity information, which seriously damages the interests of financial institutions and customers, and also poses a threat to the stability of the financial market.

[0053] See also Figure 1 The existing face verification process generally involves collecting face image data through a camera, and then performing liveness detection and face comparison on the face image data in sequence to obtain the identity verification result. Figure 2 The existing facial image data used for facial authentication is generally obtained by irradiating light through a lens onto a sensor, and the camera captures photoelectrons through a photosensitive element to form a Raw image (i.e., the original lossless image). The original lossless image is then split into single-channel images of three primary colors and interpolated to obtain three primary color layers. The three primary color layers are then processed for image signals (such as denoising, white balance, de-mosaicing, etc.) to generate a full-color image (such as a JPG image, a PNG image, etc.), and then the full-color image (i.e., the facial image data) is output. The processed facial image data will be sequentially processed through the liveness detection, facial comparison, and other links of the facial authentication process to complete the verification of identity information. The AIGC technology can replace the facial image data with forged image data to enter the facial authentication process, thereby verifying the identity information that has been illegally passed. Therefore, the forged information generated by the existing AIGC technology reduces financial security.

[0054] The present application provides a solution, using the original lossless image as the input of face authentication. On the one hand, since the original lossless image is the original image formed by the camera capturing photoelectrons through a photosensitive element (such as CCD, CMOS, etc.), rather than the full-color image output by processing the image signal of the original lossless image formed by the camera in the prior art, the original lossless image is encapsulated inside the camera, and it is difficult for attackers to attack the hardware bottom layer to tamper with the original lossless image. On the other hand, the current AIGC model is basically obtained by training with full-color images. Therefore, if the AIGC model is used to forge the original lossless image, it is necessary to invest a lot of technical costs and time costs to completely retrain the new model. In addition, since the original lossless image is the image data of the hardware bottom layer, the number of original lossless images on the market is small, and the training of the AIGC model requires an extremely large number of training samples. Therefore, in the case of insufficient training samples, the trained new model may also be completely unavailable, and the investment cost is much higher than the counterfeiting income. In addition, for the existing face verification process, this application does not make any other changes to the main process of the existing face verification process except replacing the input image from a full-color image to an original lossless image, so it has good compatibility with the existing face verification process. Therefore, this application performs face verification based on the original lossless image, which can effectively resist the attack of forged information generated by the existing AIGC technology and improve financial security.

[0055] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device, a personal verification device, a server, etc. that can realize the above functions. The following takes the personal verification device as an example to illustrate this embodiment and the following embodiments.

[0056] Based on this, the present application embodiment provides an identity verification method, referring to Figure 3 , Figure 3 This is a flowchart of the first embodiment of the identity verification method of this application.

[0057] In this embodiment, the identity verification method includes steps S10 to S40:

[0058] Step S10, in response to the face verification instruction, obtaining the original lossless image captured by the camera;

[0059] It should be noted that the execution subject of this embodiment is the identity verification device, and the face verification instruction is a control instruction that instructs the identity verification device to execute a predetermined face verification process, and the face verification instruction can be issued by the user, or issued after the identity verification device meets a predetermined trigger condition. The camera can be a camera set on the identity verification device, or a camera connected to the identity verification device through wired communication or wireless communication.

[0060] In addition, it should be noted that the original lossless image is the original data of the light source signal captured by the optical sensor of the camera converted into a digital signal. It can be understood that the format of the original lossless image is determined by the color filter array of the camera. For example, if the color filter array of the camera is a Bayer array, the original lossless image is stored in the format of the Bayer array. For example, the original lossless image can also be stored in the format of the Foveon X3 array, the X-Trans array, and other non-standard arrays.

[0061] This embodiment can obtain the original lossless image captured by the camera in response to the face verification instruction after receiving the face verification instruction sent by the user or the face verification instruction triggered by the identity verification device satisfying a predetermined trigger condition. Figure 4 , Figure 4 Schematic diagram of the array structure of the Bayer array involved in the embodiment of the present application. Figure 4 Four types of Bayer arrays are shown, namely RGGB, BGGR, GRBG and GBRG. Figure 5 This is an example diagram of the original lossless image and the converted full-color image involved in the embodiments of the present application. Figure 5 The left image is the original lossless image in Bayer format (i.e., each pixel contains only one of RGB, which is a single-channel image), and the right image is the converted full-color image (i.e., each pixel contains RGB information).

[0062] Step S20, performing face verification processing based on the original lossless image to obtain an identity verification result.

[0063] It should be noted that the facial authentication processing is a pre-set processing operation for identity verification based on the face, and the facial authentication processing includes at least a liveness detection operation and a face comparison operation. The liveness detection operation is an operation for detecting whether the person in the original lossless image is alive, and the face comparison operation is an operation for comparing whether the person in the original lossless image is the target person.

[0064] As an example, this embodiment can directly perform liveness detection on the original lossless image to obtain a liveness detection result. After the liveness detection result is that the liveness detection passes, the original lossless image is subjected to face comparison to obtain an identity verification result.

[0065] As another example, in order to improve the image quality of the original lossless image so as to facilitate processing and recognition in the subsequent face identification process, the present embodiment may also perform predetermined image processing on the original lossless image to obtain a preprocessed image. The predetermined image processing may include one or more processing operations to improve image quality, such as splitting the three primary color channels, interpolating the split single channel image, noise filtering, normalization, detail enhancement, and illumination compensation. Then, liveness detection is performed on the preprocessed image to obtain a liveness detection result. After the liveness detection result is that the liveness detection passes, face comparison is performed on the preprocessed image to obtain an identity verification result.

[0066] In a feasible implementation, step S20 may include steps S21 to S23:

[0067] Step S21, performing predetermined image processing on the original lossless image to obtain a preprocessed image;

[0068] Step S22, performing liveness detection on the preprocessed image to obtain a liveness detection result;

[0069] Step S23, after the liveness detection result is that the liveness detection is passed, face comparison is performed on the pre-processed image to obtain an identity verification result.

[0070] It should be noted that the predetermined image processing is a pre-set processing operation to improve image quality, so as to facilitate subsequent image analysis and recognition in the process of living body detection and face comparison, and may include one or more processing operations to improve image quality, such as splitting the three primary color channels, interpolating the split single channel images, noise filtering, normalization, detail enhancement, and illumination compensation.

[0071] In addition, it should be noted that the liveness detection is an operation of detecting whether the person in the original lossless image is a real live person. The face comparison is an operation of comparing the face image in the original lossless image with the face image of the target person in the preset database.

[0072] In this embodiment, the original lossless image can be processed by at least one image processing operation such as splitting the three primary color channels, interpolating the split single channel image, noise filtering, normalization, detail enhancement, and illumination compensation to obtain a preprocessed image. Then, the preprocessed image can be subjected to liveness detection by at least one of liveness detection operations such as interactive liveness detection, light liveness detection, silent liveness detection, and 3D face liveness detection to obtain a liveness detection result. The liveness detection result may include liveness detection passing or liveness detection failing. It can be understood that this embodiment does not limit the specific implementation method of the liveness detection. If the liveness detection result is that the liveness detection fails, the identity verification result is that the identity verification fails. After the liveness detection result is that the liveness detection passes, the preprocessed image is subjected to face comparison to obtain an identity verification result. In this embodiment, the target facial information of the target person can be obtained, and then the preprocessed image and the target facial information are compared. If the preprocessed image matches the target facial information, the identity verification passing can be used as the identity verification result. If the pre-processed image does not match the target facial information, the identity verification failure may be taken as the identity verification result.

[0073] In a feasible implementation, step S21 may include steps A10 to A20:

[0074] Step A10, obtaining the color filter array of the camera;

[0075] Step A20, splitting the original lossless image according to the arrangement positions of the primary colors in the color filter array to obtain primary color channel images as preprocessed images.

[0076] Since the original lossless image of the Bayer array is a single-channel format image, in order to improve the image quality of this type of single-channel format image, so as to facilitate the processing and recognition in the subsequent face identification process. Since different cameras may use different color filter arrays, this embodiment can obtain the color filter array of the camera, and then split the original lossless image according to the arrangement position of each primary color in the color filter array to obtain each primary color channel map as a preprocessed image. For example, Figure 6As shown, this embodiment can split the original lossless image into primary color channel images of three primary colors: red, green, and blue. Since the Bayer array is composed of a 2*2 color array, this 2*2 array consists of two green, one red, and one blue. Therefore, this embodiment can further split the green color to obtain two green primary color channel images. In this way, the original lossless image in the Bayer format with a dimension size of (H, W, 1) is split into four primary color channel images with a dimension size of (H / 2, W / 2, 4) (one red, one blue, and two green primary color channel images).

[0077] In this embodiment, the color filter array of the camera is obtained; the original lossless image is split according to the arrangement position of each primary color in the color filter array, and each primary color channel map is obtained as a preprocessed image. On the one hand, by splitting the original lossless image, the separation of different primary color channels is achieved, which helps to reflect the distribution and change of colors in the original lossless image, and in the subsequent process of processing the primary color channel map of each primary color separately, more refined and targeted features can be extracted, thereby improving the image quality, which is beneficial to subsequent operations such as living body detection and face comparison. On the other hand, the original lossless image is split with the help of the color filter array of the camera. Compared with the method of traversing the image and identifying the position of each primary color and then splitting it in the prior art, since the position of each primary color in the original lossless image in this embodiment corresponds to the color filter array of the camera, the efficiency of splitting the original lossless image can be effectively improved.

[0078] The first embodiment of the present application provides an identity verification method, which obtains the original lossless image formed by the camera shooting in response to the face verification instruction, and then performs face verification processing based on the original lossless image to obtain the identity verification result. In this application, the original lossless image is used as the input of face verification. On the one hand, since the original lossless image is the original image formed by the camera capturing photoelectrons through a photosensitive element (such as CCD, CMOS, etc.), rather than the full-color image output after image signal processing of the original lossless image formed by the camera in the prior art, the original lossless image is encapsulated inside the camera, and it is difficult for an attacker to attack the hardware bottom layer to tamper with the original lossless image. On the other hand, the current AIGC models are basically trained based on full-color images. Therefore, if the AIGC model is used to forge the original lossless image, a large amount of technical and time costs will be required to completely retrain the new model. In addition, since the original lossless image is the image data of the hardware bottom layer, the number of original lossless images on the market is relatively small, and the training of the AIGC model requires an extremely large number of training samples. Therefore, in the case of insufficient training samples, the trained new model may be completely unusable, and the investment cost is much higher than the profit from counterfeiting. In addition, for the existing face verification process, this embodiment does not make any other changes to the main process of the existing face verification process except replacing the input image from the full-color image to the original lossless image, so it has good compatibility with the existing face verification process. Therefore, this embodiment performs face verification based on the original lossless image, which can effectively resist the attack of forged information generated by the existing AIGC technology and improve financial security.

[0079] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following. Figure 7 , step S22 includes steps A10 to A20:

[0080] Step B10, inputting the preprocessed image into a preset living body detection model to obtain credibility;

[0081] Step B20: after the liveness credibility is greater than a preset credibility threshold, the liveness detection is passed as the liveness detection result.

[0082] It should be noted that the preset liveness detection model is a pre-trained neural network model for liveness detection, such as a convolutional neural network model, a recurrent neural network model, a generative adversarial network, etc. It is understandable that the specific selection of the preset liveness detection model can be determined according to the liveness detection method adopted.

[0083] In this embodiment, after the pre-processed image is input into a preset liveness detection model, the credibility of the output of the preset liveness detection model can be obtained, and the credibility represents the credibility of the person in the pre-processed image as a live body. When the liveness credibility is greater than the preset credibility threshold, it means that the credibility of the person in the pre-processed image as a live body is high, and the liveness detection pass can be used as the liveness detection result. If the liveness credibility is not greater than the preset credibility threshold, it means that the credibility of the person in the pre-processed image as a live body is low, and the liveness detection failure can be used as the liveness detection result, and the identity verification result is also the identity verification failure.

[0084] In a feasible embodiment, steps C10 to C20 are included before step B10:

[0085] Step C10, obtaining a first training sample, wherein the first training sample comprises a first image sample in an original lossless format;

[0086] Step C20: training a first initial model based on the first image sample until the first initial model converges to obtain a preset living body detection model.

[0087] It should be noted that the first training sample includes a first image sample in an original lossless format, that is, the first image sample is an image in an original lossless format. The first initial model is an untrained liveness detection model, which can be an untrained convolutional neural network model, a recurrent neural network model, a generative adversarial network or other neural network model.

[0088] In this embodiment, a first training sample is obtained, and the first training sample includes a first image sample in an original lossless format. Then, the first initial model is trained based on the first image sample until the first initial model converges (such as the loss function of the first initial model is minimized or the number of iterations of the first initial model reaches the maximum number of iterations), and the converged first initial model can be used as the preset liveness detection model. In this embodiment, the first initial model is trained by the first image sample in the original lossless format, so that the preset liveness detection model can have a higher detection accuracy for the original lossless image.

[0089] In the second embodiment of the present application, the credibility is obtained by inputting the pre-processed image into a preset liveness detection model; after the liveness credibility is greater than the preset credibility threshold, the liveness detection is passed as the liveness detection result. Therefore, this embodiment uses the original lossless image after image processing that is difficult to be attacked and tampered with, and the features that are difficult to forge in the existing AIGC model to perform liveness detection operations, effectively improving the accuracy of liveness detection results, thereby improving the accuracy of face verification.

[0090] Based on the first embodiment of the present application, in the third embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following. Figure 8 , step S23 includes steps D10 to D30:

[0091] Step D10, obtaining target facial information of the target person;

[0092] Step D20, inputting the preprocessed image and the target facial information into a preset face comparison model to obtain a matching degree between the preprocessed image and the target facial information;

[0093] Step D30, after the matching degree is greater than a preset matching threshold, the identity verification is passed as the identity verification result.

[0094] It should be noted that the target person is a person who wishes to undergo facial verification, and the target facial information may be a facial image of the target person, or facial features obtained by extracting features from the facial image. It is understandable that the facial image of the target person may be a full-color image, or may be an image in an original lossless format.

[0095] This embodiment can obtain the target facial information of the target person from a preset database (such as a local database or a cloud database), and then input the preprocessed image and the target facial information into a preset face comparison model to obtain the matching degree between the preprocessed image and the target facial information, wherein the matching degree represents the degree of similarity between the person in the preprocessed image and the target person. When the matching degree is greater than a preset matching threshold, it means that the similarity between the person in the preprocessed image and the target person is high, and the identity verification pass can be used as the identity verification result. If the matching degree is not greater than the preset matching threshold, it means that the similarity between the person in the preprocessed image and the target person is low, and the identity verification failure can be used as the identity verification result.

[0096] In a feasible embodiment, before step D20, steps E10 to E20 are included:

[0097] Step E10, obtaining a second training sample, wherein the second training sample comprises a second image sample in an original lossless format;

[0098] Step E20: training a second initial model based on the second image sample until the second initial model converges to obtain a preset face comparison model.

[0099] It should be noted that the second training sample includes a second image sample in an original lossless format, that is, the second image sample is an image in an original lossless format. The second initial model is an untrained face comparison model, which can be an untrained convolutional neural network model, a recurrent neural network model, a generative adversarial network or other neural network model.

[0100] In this embodiment, the second training sample is obtained, and the second training sample includes a second image sample in an original lossless format. Then, the second initial model is trained based on the second image sample until the second initial model converges (such as the loss function of the second initial model is minimized or the number of iterations of the second initial model reaches the maximum number of iterations), and the converged second initial model can be used as the preset face comparison model. In this embodiment, the second initial model is trained by the second image sample in the original lossless format, so that the preset face comparison model can have a higher recognition accuracy for the original lossless image.

[0101] In the third embodiment of the present application, the target facial information of the target person is obtained; the pre-processed image and the target facial information are input into a preset face comparison model to obtain the matching degree between the pre-processed image and the target facial information; after the matching degree is greater than a preset matching threshold, the identity verification is passed as the identity verification result. Therefore, this embodiment uses the original lossless image after image processing that is difficult to be attacked and tampered with, and the features that are difficult to forge in the existing AIGC model to perform face comparison operations, which effectively improves the accuracy of the face comparison results, thereby improving the accuracy of face verification.

[0102] See also Fig. 9 , Fig. 9 This is a scene comparison diagram of face verification involved in the embodiment of the present application. Fig. 9 As shown, Fig. 9The difference between the identity verification method of the present application and the existing face verification method is shown. The present application performs face verification directly when the camera captures photoelectrons to form the original lossless image, rather than performing face verification in the image data processed by the camera back end. Therefore, the face verification operation of the present application is performed at the front end of the camera. Since the original lossless image is directly formed inside the hardware of the camera, it is difficult for attackers to use AIGC technology to attack the underlying hardware, thereby isolating the impact of AIGC attacks. Moreover, the current AIGC model is basically obtained by training with full-color images, so its output is a full-color image, and the original lossless image cannot be directly forged. If the AIGC model is used to forge the original lossless image, it is necessary to invest a lot of technical costs and time costs to completely retrain the new model. Moreover, since the original lossless image is the image data of the underlying hardware, the number of original lossless images on the market is small, and the training of the AIGC model requires an extremely large number of training samples to be possible. Therefore, in the case of insufficient training samples, the trained new model may also be completely unavailable, and the investment cost is much higher than the counterfeiting income.

[0103] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the identity verification method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0104] The present application provides an identity verification device, which includes: at least one processor; and a memory that is communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the identity verification method in the above-mentioned embodiment one.

[0105] Reference below Fig.10 , which shows a schematic diagram of the structure of an identity verification device suitable for implementing the embodiment of the present application. The identity verification device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle controllers), etc., and fixed terminals such as digital TVs, desktop computers, etc. Fig.10 The identity verification device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0106] like Fig.10As shown, the identity verification device may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the identity verification device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can allow the identity verification device to communicate wirelessly or wired with other devices to exchange data. Although the figure shows an identity verification device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or provided alternatively.

[0107] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0108] The identity verification device provided by the present application adopts the identity verification method in the above embodiment, which can solve the technical problem that the forged information generated by the existing AIGC technology reduces financial security. Compared with the prior art, the beneficial effects of the identity verification device provided by the present application are the same as the beneficial effects of the identity verification method provided by the above embodiment, and the other technical features in the identity verification device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0109] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0110] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0111] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, wherein the computer-readable program instructions are used to execute the identity verification method in the above-mentioned embodiment.

[0112] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.

[0113] The computer-readable storage medium may be included in the identity verification device; or may exist independently without being assembled into the identity verification device.

[0114] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by the identity verification device, the identity verification device: responds to the face verification instruction, obtains the original lossless image taken by the camera, and then performs face verification processing based on the original lossless image to obtain the identity verification result.

[0115] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0116] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0117] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0118] The readable storage medium provided in this application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned identity verification method, and can solve the technical problem that the forged information generated by the existing AIGC technology reduces financial security. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the identity verification method provided in the above-mentioned embodiment, and will not be repeated here.

[0119] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned identity verification method when executed by a processor.

[0120] The computer program product provided by this application can solve the technical problem that the forged information generated by the existing AIGC technology reduces financial security. Compared with the prior art, the beneficial effects of the computer program product provided by this application are the same as the beneficial effects of the identity verification method provided by the above embodiment, which will not be repeated here.

[0121] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. An identity verification method, characterized in that: The identity verification method includes: In response to a face verification instruction, an original lossless image captured by a camera is obtained; Face verification processing is performed based on the original lossless image to obtain an identity verification result.

2. The method according to claim 1, characterized in that The step of performing face verification processing based on the original lossless image to obtain an identity verification result comprises: Performing predetermined image processing on the original lossless image to obtain a preprocessed image; Performing liveness detection on the preprocessed image to obtain a liveness detection result; After the liveness detection result is that the liveness detection is passed, face comparison is performed on the pre-processed image to obtain an identity verification result.

3. The method according to claim 2, characterized in that The step of performing liveness detection on the original lossless image to obtain a liveness detection result comprises: Inputting the preprocessed image into a preset living body detection model to obtain credibility; After the liveness credibility is greater than a preset credibility threshold, the liveness detection passing is taken as the liveness detection result.

4. The method according to claim 3, characterized in that Before the step of inputting the preprocessed image into a preset living body detection model to obtain the credibility, the method includes: Acquire a first training sample, wherein the first training sample comprises a first image sample in an original lossless format; The first initial model is trained based on the first image sample until the first initial model converges to obtain a preset living body detection model.

5. The method according to claim 2, characterized in that The step of performing face comparison on the original lossless image to obtain an identity verification result includes: Obtain target facial information of the target person; Inputting the preprocessed image and the target facial information into a preset face comparison model to obtain a matching degree between the preprocessed image and the target facial information; After the matching degree is greater than a preset matching threshold, the identity verification is passed as the identity verification result.

6. The method according to claim 5, characterized in that The step of inputting the preprocessed image and the target facial information into a preset face comparison model to obtain a matching degree between the preprocessed image and the target facial information comprises: Acquire a second training sample, wherein the second training sample comprises a second image sample in an original lossless format; The second initial model is trained based on the second image sample until the second initial model converges to obtain a preset face comparison model.

7. The method according to any one of claims 2 to 6, characterized in that The step of performing predetermined image processing on the original lossless image to obtain a preprocessed image comprises: Obtaining a color filter array of the camera; According to the arrangement positions of the primary colors in the color filter array, the original lossless image is split to obtain the primary color channel images as preprocessed images.

8. An identity verification device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the identity verification method according to any one of claims 1 to 7.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the identity verification method according to any one of claims 1 to 7 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the identity verification method according to any one of claims 1 to 7 are implemented.

Images