Method, device and equipment for identifying disguise of photo in identification certificate and storage medium

Abstract

The application relates to a method and device for identifying certificate photo camouflage in a certificate, equipment and a storage medium, which can effectively identify a certificate with a large-head sticker covering certificate photo camouflage and improve the accuracy of certificate authenticity identification. The method comprises the following steps: obtaining a certificate image, wherein the certificate image comprises a certificate photo; determining a face area in the certificate photo of the certificate image; obtaining a certificate photo sub-image from the certificate image according to the face area, wherein the certificate photo sub-image comprises a boundary of the certificate photo; obtaining a feature of the certificate photo in the certificate photo sub-image; and judging whether the certificate photo camouflage exists in the certificate image according to the feature. The application can effectively identify a certificate with a large-head sticker covering certificate photo camouflage and improve the accuracy of certificate authenticity identification.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method, apparatus, device, and storage medium for identifying spoofed ID photos in ID documents. Background Technology

[0002] With the widespread adoption of the internet and the development of technology, many services in industries such as securities, banking, and insurance can increasingly be conducted on terminal devices (such as VTMs and mobile phones), with online banking and mobile banking being the most common examples. These services, according to regulations, all require the customer to present their valid personal identification document for processing.

[0003] Traditionally, transactions at branch counters require customers to present their valid ID in person for on-site verification by staff. Moving these services to mobile devices raises the main challenge of verifying customer identity: confirming that the customer is indeed presenting their valid ID. This process involves two stages of verification: ID card information verification and verification of identity. For ID card verification, a common practice is to have customers take photos of the front and back of their ID card and their face, which are then sent to back-end staff for manual review (for all ID cards).

[0004] After ID card data collection, it's necessary to verify the authenticity of the ID card. Currently, some criminals are using photo stickers to cover the ID photo area to forge ID cards. How to effectively identify ID cards disguised with photo stickers and improve the accuracy of ID card authenticity verification is a problem that needs to be solved. Summary of the Invention

[0005] This application provides a method, apparatus, device, and storage medium for identifying disguised ID photos in ID documents, which can effectively identify ID documents disguised by covering up ID photos with photo stickers, thereby improving the accuracy of ID document authenticity identification.

[0006] In a first aspect, embodiments of this application provide a method for identifying forged ID photos in identification documents, including:

[0007] Obtain an image of an identification document, wherein the image of the identification document includes a photograph of the document;

[0008] Identify the facial region in the ID photo;

[0009] Based on the face region, an ID photo sub-image is extracted from the ID image, wherein the ID photo sub-image includes the boundaries of the ID photo;

[0010] Obtain the features of the ID photo in the ID photo sub-image;

[0011] Based on the aforementioned features, it is determined whether the document image contains a fake ID photo.

[0012] Optionally, based on the face region, an ID photo sub-image is extracted from the ID image, including:

[0013] The face region is taken as the central region, and an area of ​​a predetermined size is expanded outward from the central region. The area obtained after expansion is taken as the ID photo image.

[0014] Optionally, the feature is used to indicate whether there is a pasting effect between the ID photo and the ID document in the ID photo sub-image.

[0015] Optionally, determining whether there is an ID photo disguise in the ID image based on the features includes:

[0016] If it is determined that the feature indicates that there is a pasting effect between the ID photo and the ID document in the ID photo sub-image, then it is determined that there is an ID photo disguise in the ID photo image;

[0017] If the feature indicates that there is a pasting effect between the ID photo and the ID document in the ID photo sub-image, then it is determined that there is no ID photo spoofing in the ID photo image.

[0018] Optionally, obtaining the features of the ID photo in the ID photo sub-image includes:

[0019] The ID photo image is input into the recognition model to obtain the features of the ID photo in the ID photo image output by the recognition model;

[0020] The recognition model is obtained by training a preset model using a set of sample document images. The set of sample document images includes the correspondence between the sample document images and features. The features are used to indicate whether there is a pasting effect between the document photo and the document in the document image.

[0021] Optionally, features of the face region in the sample document image are removed;

[0022] Before calculating the features of the document in the document photo sub-image, the method further includes:

[0023] Remove features from the facial region in the document image.

[0024] Optionally, the ratio of the area occupied by the region of the set size to the area of ​​the document image is a set value.

[0025] Secondly, embodiments of this application provide an apparatus for identifying forged ID photos in identification documents, comprising:

[0026] An acquisition module is used to acquire an ID card image, wherein the ID card image includes an ID card photograph;

[0027] The determination module is used to determine the face region in the ID photo of the ID document image;

[0028] The cropping module is used to crop an ID photo sub-image from the ID image based on the face region, wherein the ID photo sub-image includes the boundary of the ID photo;

[0029] The recognition module is used to obtain the features of the ID photo in the ID photo sub-image;

[0030] The judgment module is used to determine whether there is an ID photo disguise in the ID image based on the features.

[0031] Thirdly, embodiments of this application provide an electronic device, including: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus;

[0032] The memory is used to store computer programs;

[0033] The processor is used to execute the program stored in the memory to implement the method for identifying the forgery of ID photos in ID documents as described in the first aspect.

[0034] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the method for identifying the forgery of ID photos in a document as described in the first aspect.

[0035] Compared with the prior art, the above-mentioned technical solution provided in this application has the following advantages: The method provided in this application identifies the face region in the ID photo of the ID image, extracts the ID photo sub-image from the ID image based on the face region, and obtains the features of the ID photo in the ID photo sub-image. This enables the effective identification of whether there is a fake ID photo such as a headshot in the ID image based on the features, thereby improving the accuracy of ID authenticity identification. Attached Figure Description

[0036] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0037] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a schematic diagram of the method for identifying the disguise of ID photos in documents in the embodiments of this application;

[0039] Figure 2 This is a schematic diagram showing the expansion of the face region in an embodiment of this application;

[0040] Figure 3 This is a schematic diagram of the device structure for identifying the disguise of ID photos in documents in the embodiments of this application;

[0041] Figure 4 This is a schematic diagram of the structure of the electronic device in the embodiments of this application. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0043] In this embodiment of the application, in order to solve the problem that the authenticity of the document cannot be effectively identified after a photo disguised as an ID photo is used in the document, resulting in identification errors, a method for identifying the disguise of ID photos in documents is proposed. This method can be applied to any electronic device, which can be a terminal device or a server.

[0044] like Figure 1 As shown, the specific process for identifying fake ID photos on documents is as follows:

[0045] Step 101: Obtain an ID image, wherein the ID image includes an ID photo.

[0046] If the terminal device identifies the spoofing of the ID photo in the ID document, then the terminal device acquires the ID document image. If the server identifies the spoofing of the ID photo in the ID document, then the terminal device acquires the ID document image and transmits it to the server.

[0047] In one specific embodiment, the process of capturing an ID card image using a mobile phone as an example is as follows: The mobile client displays a guide box on the screen for capturing the ID card image. The mobile client prompts (voice or text prompts) the user to face the front of the ID card towards the mobile phone screen and place the ID card within the guide box, adjusting the shooting posture to ensure that the brightness, blur, borders, etc., meet the capture requirements. The mobile client captures the image within the guide box to obtain the front image of the ID card. The process of capturing the back image of the ID card is similar and will not be repeated here.

[0048] Among them, the document image can be an image of the side of the document that includes the ID photo, such as the front image of an ID card.

[0049] Step 102: Determine the face region in the ID photo within the ID image.

[0050] Any facial feature recognition or localization algorithm can be used to determine the facial region in the document image. For example, a traditional face detection method based on histogram of oriented gradients (HOG) or a face detection method based on convolutional neural networks (CNN) can be used to locate the facial region in the document image.

[0051] Step 103: Based on the face region in the ID card image, extract the ID photo sub-image from the ID card image, wherein the ID photo sub-image includes the boundary of the ID photo.

[0052] In one specific embodiment, the face region in the ID card image is used as the central region, and an area of ​​a predetermined size is expanded outward from this central region. The expanded area is then used as the ID card photo sub-image. For example, as... Figure 2 As shown, the shadow area is obtained after the face area is expanded outward. This shadow area can cover the entire ID photo area and includes the adjacent area of ​​the ID photo area.

[0053] The ratio of the area occupied by the set size region to the area of ​​the document image is a set value.

[0054] Specifically, after identifying the face region, the bounding rectangle of this face region is expanded outwards. The expansion is done by a first proportion along the horizontal axis and a second proportion along the vertical axis. The expanded area is used as the ID photo sub-image. The first proportion is determined by statistically analyzing the ratio of the length of the face region on the horizontal axis to the length of the ID photo region on the horizontal axis in the sample ID image, and the average of these ratios is taken as the first proportion. Similarly, the second proportion is determined by statistically analyzing the ratio of the length of the face region on the vertical axis to the length of the ID photo region on the vertical axis in the sample ID image, and the average of these ratios is taken as the second proportion. In the experiment, the first proportion was found to be 1.3, and the second proportion was 1.6.

[0055] The adjacent area occupies a very small proportion in the ID photo sub-image; it is only necessary to be able to completely contain the boundary of the ID photo area within the ID photo image.

[0056] Step 104: Obtain the features of the ID photo in the ID photo sub-image.

[0057] Among them, the feature is used to indicate whether there is a pasting effect between the ID photo and the ID document in the ID photo sub-image.

[0058] In one specific embodiment, the process of identifying the features of the ID photo in the ID photo sub-image is as follows: the ID photo sub-image is input into the recognition model, and the features of the ID photo in the ID photo sub-image output by the recognition model are obtained; wherein, the recognition model is obtained by training a preset model with a set of sample ID photos, wherein the set of sample ID photos includes the correspondence between sample ID photos and features, and the features are used to indicate whether there is a pasting effect between the ID photo and the ID in the ID photo.

[0059] The sample document images include multiple genuine document images and multiple document images forged using photo stickers. The features of the document photos in each sample document image are labeled. The set of sample document images is used to train the preset model until it can correctly determine whether the document image contains a photo sticker and the document being pasted together.

[0060] The preset model can be a residual neural network model, such as ResNet50 or ResNet18, or other network models, such as GoogleNet or DenseNet.

[0061] To improve the recognition accuracy of the trained model, facial features are removed from the sample ID images before training. For example, an occlusion template is used to cover the facial areas in the sample ID images. This occlusion template can be a single-color occlusion image, such as a black or white image. Occlusion removes interference from facial features during training. Compared to training the model using the entire captured image, this method reduces interference factors and improves recognition accuracy.

[0062] Furthermore, before obtaining the features of the ID photo in the ID photo sub-image, the features of the face region in the ID photo image or ID photo sub-image are removed. That is, the face region in the ID photo sub-image is covered by an occlusion template to reduce interference and improve recognition accuracy.

[0063] Similarly, before locating the face region, features that are clearly not in the ID photo area, such as text areas, can be occluded to reduce the complexity of face localization and improve its accuracy.

[0064] Experiments show that without occlusion of the face area, the model training and feature recognition focus on the face. After occlusion of the face features, the model training and feature recognition focus on the boundaries of the ID photo, which greatly improves the recognition accuracy of the model.

[0065] Step 105: Based on this feature, determine whether there is any spoofing of the ID photo in the ID image.

[0066] Specifically, if it is determined that the feature indicates that there is a pasting effect between the ID photo and the ID document in the ID photo sub-image, then it is determined that there is ID photo spoofing in the ID photo image; if it is determined that the feature indicates that there is a pasting effect between the ID photo and the ID document in the ID photo sub-image, then it is determined that there is no ID photo spoofing in the ID photo image.

[0067] After determining that there is no fake ID photo in the ID image, further detection can be performed on the ID image.

[0068] Based on product requirements, if it is determined that an ID photo is faked, the information of the ID can be added to a blacklist, or a new ID image can be collected.

[0069] In this embodiment, by identifying the facial region in the ID photo of an ID document image, and extracting a sub-image of the ID photo from the ID document image based on the facial region, the features of the ID photo in the sub-image are obtained. This allows for effective identification of whether the ID document image contains a fake ID photo with a headshot or other disguised image, thus improving the accuracy of ID document authenticity verification. Furthermore, by using a method of occluding the facial region, the accuracy of ID document authenticity verification is further improved.

[0070] Based on the same concept, this application provides a device for identifying forged ID photos in identification documents. The specific implementation of this device can be found in the description of the method embodiments section; repeated details will not be repeated here. Figure 3 As shown, the device mainly includes:

[0071] The acquisition module 301 is used to acquire an ID card image, wherein the ID card image includes an ID card photo;

[0072] The determining module 302 is used to determine the face region in the ID photo of the ID document image;

[0073] The cropping module 303 is used to crop an ID photo sub-image from the ID image based on the face region, wherein the ID photo sub-image includes the boundary of the ID photo;

[0074] The recognition module 304 is used to obtain the features of the ID photo in the ID photo sub-image;

[0075] The judgment module 305 is used to determine whether there is an ID photo disguise in the ID image based on the features.

[0076] Specifically, the cropping module 303 is used to: take the face region as the central region, expand the central region outwards by a predetermined size, and use the expanded region as the ID photo image. The ratio of the area occupied by the predetermined size region to the area of ​​the ID photo is a predetermined value.

[0077] Specifically, the feature is used to indicate whether there is a pasting effect between the ID photo and the ID document in the ID photo sub-image.

[0078] Specifically, if the determination module 305 determines that the feature indicates that there is a pasting effect between the ID photo and the ID document in the ID photo sub-image, then it determines that there is ID photo spoofing in the ID photo image; if the determination module 305 determines that the feature indicates that there is a pasting effect between the ID photo and the ID document in the ID photo sub-image, then it determines that there is no ID photo spoofing in the ID photo image.

[0079] Specifically, the recognition module 304 is used to input the ID photo sub-image into the recognition model to obtain the features of the ID photo in the ID photo sub-image output by the recognition model; wherein, the recognition model is obtained by training a preset model using a set of sample ID photos, wherein the set of sample ID photos includes the correspondence between sample ID photos and features, and the features are used to indicate whether there is a pasting effect between the ID photo and the ID in the ID photo.

[0080] In this process, features of the facial region in the sample document image are removed.

[0081] Before recognizing the features of the ID photo in the ID photo sub-image, the recognition module 304 removes the features of the face region in the ID photo sub-image.

[0082] Based on the same concept, this application also provides an electronic device, such as... Figure 4 As shown, the electronic device mainly includes: a processor 401, a communication interface 402, a memory 403, and a communication bus 404. The processor 401, communication interface 402, and memory 403 communicate with each other via the communication bus 404. The memory 403 stores a program executable by the processor 401. The processor 401 executes the program stored in the memory 403 to perform the following steps: acquiring an ID image, wherein the ID image includes an ID photo; determining the face region in the ID photo; based on the face region, extracting an ID photo sub-image from the ID image, wherein the ID photo sub-image includes the boundaries of the ID photo; acquiring the features of the ID photo in the ID photo sub-image; and determining whether there is ID photo spoofing in the ID image based on the features.

[0083] The communication bus 404 mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus 404 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0084] Communication interface 402 is used for communication between the above-mentioned electronic device and other devices.

[0085] The memory 403 may include random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor 401.

[0086] The processor 401 mentioned above can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc., or a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0087] In another embodiment of this application, a computer-readable storage medium is also provided, which stores a computer program that, when run on a computer, causes the computer to perform the method for identifying the forgery of ID photos in the above embodiments.

[0088] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. This computer program product includes one or more computer instructions. When these computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another, for example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape, etc.), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0089] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0090] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for identifying forged ID photos in identification documents, characterized in that, include: Obtain an image of an identification document, wherein the image of the identification document includes a photograph of the document; Identify the facial region in the ID photo; Based on the face region, an ID photo sub-image is extracted from the ID image, wherein the ID photo sub-image includes the boundaries of the ID photo; Obtain the features of the ID photo in the ID photo sub-image; Based on the aforementioned features, it is determined whether the document image contains a disguised ID photo. Based on the facial region, an ID photo sub-image is extracted from the ID image, including: The face region is taken as the central region, and an area of ​​a predetermined size is expanded outward from the central region. The area obtained after expansion is taken as the ID photo image.

2. The method for identifying forged ID photos in identification documents according to claim 1, characterized in that, The feature is used to indicate whether there is a pasting effect between the ID photo and the ID document in the ID photo sub-image.

3. The method for identifying disguised ID photos in identification documents according to claim 2, characterized in that, Determining whether the ID photo in the ID image is a fake ID photo based on the aforementioned features includes: If it is determined that the feature indicates that there is a pasting effect between the ID photo and the ID document in the ID photo sub-image, then it is determined that there is an ID photo disguise in the ID photo image; If it is determined that the feature indicates that there is no pasting effect between the ID photo and the ID document in the ID photo sub-image, then it is determined that there is no ID photo spoofing in the ID photo image.

4. The method for identifying forged ID photos in identification documents according to claim 2, characterized in that, Obtaining the features of the ID photo in the ID photo sub-image includes: The ID photo image is input into the recognition model to obtain the features of the ID photo in the ID photo image output by the recognition model; The recognition model is obtained by training a preset model using a set of sample document images. The set of sample document images includes the correspondence between the sample document images and features. The features are used to indicate whether there is a pasting effect between the document photo and the document in the document image.

5. The method for identifying forged ID photos in identification documents according to claim 4, characterized in that, Features of the facial region in the sample document image were removed; Before obtaining the features of the ID photo in the ID photo sub-image, the method further includes: Remove features from the face region in the document sub-image.

6. The method for identifying forged ID photos in identification documents according to claim 1, characterized in that, The ratio of the area occupied by the region of the set size to the area of ​​the document image is a set value.

7. A device for identifying forged ID photos in identification documents, characterized in that, include: An acquisition module is used to acquire an ID card image, wherein the ID card image includes an ID card photograph; The determination module is used to determine the face region in the ID photo of the ID document image; The cropping module is used to crop an ID photo image from the ID image based on the face region, wherein the ID photo image includes the boundary of the ID photo. The process of cropping the ID photo image from the ID image based on the face region includes: taking the face region as the center region, expanding the center region outward by a predetermined size, and using the expanded region as the ID photo image. The recognition module is used to obtain the features of the ID photo in the ID photo sub-image; The judgment module is used to determine whether there is an ID photo disguise in the ID image based on the features.

8. An electronic device, characterized in that, include: The processor, communication interface, memory, and communication bus are connected, with the processor, communication interface, and memory communicating with each other via the communication bus. The memory is used to store computer programs; The processor is configured to execute the program stored in the memory to implement the method for identifying the forgery of ID photos in ID documents as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method for identifying the forgery of ID photos in ID documents as described in any one of claims 1 to 6.

Images