HK40062631B - Method and device for detecting authenticity of certificate - Google Patents

Description

Methods and devices for detecting the authenticity of documents

Technical Field

This specification relates to one or more embodiments in the field of information security, and in particular to a method and apparatus for detecting the authenticity of documents.

Background Technology

In many business sectors, including finance, government affairs, and social networking, identity verification of users is often required through card-based documents such as second-generation resident ID cards and social security cards. For example, in financial scenarios such as EKYC (Know Your Customer) and AML (Anti-Money Laundering), the verification party of the financial platform requires users to collect images of their own documents on-site through terminal devices in order to verify the user's authenticity and legitimacy.

During the verification process, unauthorized users may use photocopies of genuine identification documents or other fake documents to impersonate real documents in an attempt to deceive the testing party (a situation known as a fake document attack). To ensure the reliability of user identity verification results, the authenticity of the documents provided by the user must first be guaranteed. Therefore, the testing party needs to verify the authenticity of the aforementioned card-type identification documents.

Summary of the Invention

In view of this, one or more embodiments of this specification provide a method and apparatus for detecting the authenticity of documents.

To achieve the above objectives, one or more embodiments of this specification provide the following technical solutions:

According to a first aspect of one or more embodiments of this specification, a method for detecting the authenticity of an identification document is provided, comprising:

Acquire an image of the document to be inspected, the image being captured during the rotation of the document to be inspected;

The surface information of the document to be detected is obtained based on the document image;

If the information on the surface of the document meets the preset detection conditions, the document to be tested is determined to be a genuine document.

According to a second aspect of one or more embodiments of this specification, an apparatus for detecting the authenticity of identification documents is provided, comprising:

An image acquisition unit is used to acquire an image of the document to be inspected, wherein the image is acquired during the rotation of the document to be inspected.

An information acquisition unit is used to acquire surface information of the document to be detected based on the document image;

The authenticity determination unit is used to determine that the document to be tested is an authentic document when the surface information of the document meets the preset detection conditions.

According to a third aspect of one or more embodiments of this specification, an electronic device is provided, comprising:

processor;

Memory used to store processor-executable instructions;

The processor implements the method as described in the first aspect by running the executable instructions.

According to a fourth aspect of one or more embodiments of this specification, a computer-readable storage medium is provided that stores computer instructions thereon, which, when executed by a processor, implement the steps of the method as described in the first aspect.

Attached Figure Description

Figure 1 is a schematic diagram of the architecture of a document authenticity detection system provided in an exemplary embodiment.

Figure 2 is a flowchart of a method for detecting the authenticity of a document, provided in an exemplary embodiment.

Figure 3 is a schematic diagram of the rotation process of a document to be tested provided in an exemplary embodiment.

Figure 4 is a schematic diagram of the layout of standard document content provided in an exemplary embodiment.

Figure 5 is a schematic diagram of the structure of a device provided in an exemplary embodiment.

Figure 6 is a block diagram of an exemplary embodiment of a device for detecting the authenticity of a document.

Detailed Implementation

Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with one or more embodiments of this specification. Rather, they are merely examples of apparatuses and methods consistent with some aspects of one or more embodiments of this specification as detailed in the appended claims.

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

To address forgery attacks using photocopies, two main approaches have been proposed. One approach uses data-driven deep learning to identify the document in question; however, this method has low accuracy when dealing with highly realistic color photocopies, especially cropped ones. The other approach involves acquiring two frames of images with varying brightness and analyzing the reflectivity of the document's surface to determine if it's a photocopy. However, the flash used in this method may not illuminate the document, resulting in a low success rate for identifying genuine documents.

Therefore, this specification proposes a method for detecting the authenticity of documents, which can efficiently identify fake document attacks using photocopies by detecting whether the document to be detected is genuine. This method can be applied to the document authenticity detection system shown in Figure 1. As shown in Figure 1, the system may include a network 10, several servers, such as a first server 11 and a second server 12, and several electronic devices, such as mobile phones 13, 14, and 15.

Either the first server 11 or the second server 12 can be a physical server containing an independent host, or it can be a virtual server, cloud server, etc., hosted in a host cluster. During operation, the first server 11 and the second server 12 can run server-side programs of the same or different applications to implement the relevant business functions of those applications. For example, when the first server 11 runs a program for a business platform, it can act as the server-side application for that business platform (hereinafter referred to as the business server); similarly, when the second server 12 runs a program for an identity verification platform, it can act as the server-side application for that identity verification platform (hereinafter referred to as the verification server). The method for detecting the authenticity of documents described in this specification can be applied to the aforementioned verification server. Of course, the verification server can also be integrated into the business server as a functional component running within the business server to implement the document authenticity detection function.

Mobile phones 13-15 are just one type of electronic device that users can use. In reality, users can obviously also use electronic devices such as tablets, laptops, PDAs (Personal Digital Assistants), wearable devices (such as smart glasses, smartwatches, etc.), etc., and one or more embodiments in this specification do not limit this. All of the above-mentioned electronic devices have image capture and/or video capture functions. During operation, the electronic device can run a client-side program of an application to implement the relevant business functions of that application. For example, when the electronic device runs the aforementioned authentication platform program, it can act as a client of that platform. It should be noted that the client application of the authentication platform can be pre-installed on the electronic device, allowing the client to be launched and run on the electronic device; of course, when using online "clients" such as HTML5 technology, it is not necessary to install the corresponding application on the electronic device to obtain and run the client.

In the technical solution of the embodiments described in this specification, the second server 12 can cooperate with the client running on mobile phones 13-15 to realize the authenticity detection process of the document to be tested. Specifically, the mobile phones 13-15 are used to capture images or rotated videos of the document to be tested and upload them to the second server 12, so that the second server 12 can detect the authenticity of the document to be tested using the document authenticity detection method described in the embodiments of this specification.

Alternatively, the method for detecting the authenticity of documents described in the embodiments of this specification can also be implemented by the aforementioned mobile phones 13-15 respectively. That is, the method can be run as an end-side detection scheme in the terminal device to realize the authenticity detection of the document to be detected on the end side. It can be seen that the executing entity of the method for detecting the authenticity of documents described in the embodiments of this specification can be a server such as the first server 11 or the second server 12, or a terminal device such as mobile phones 13-15. The embodiments of this specification do not limit this.

In this embodiment, the document authenticity detection system can not only detect the authenticity of documents, but also serve as an integrated functional platform for many other functions. For example, it can capture document images, capture rotating videos, display positioning markers, determine document surface information, query standard content, and execute preset operations. This specification does not limit the scope of this embodiment. Furthermore, the document authenticity detection system can also be implemented as a subsystem within an identity verification system; this specification does not limit the scope of this embodiment.

Network 10 can include various types of wired or wireless networks.

Figure 2 is a flowchart illustrating an exemplary embodiment of a method for detecting the authenticity of an identification document. As shown in Figure 2, this method can be applied to a verification party, such as the aforementioned verification server or client. The method may include the following steps:

Step 202: Obtain the image of the document to be tested, which is acquired during the rotation of the document to be tested.

In the embodiments described in this specification, the document to be detected may be a genuine document, such as a second-generation resident ID card, social security card, public transport card, or other card-type document. These documents can be rectangular, flat, or irregularly shaped. Alternatively, the user of the terminal device may use a photocopy of any genuine document to disguise it for identity verification. This photocopy can be printed from a photograph of the genuine document, and can be printed on materials such as paper or fabric, specifically as a black-and-white or color photocopy. The embodiments in this specification do not limit the shape or purpose of the document to be detected.

Typically, genuine documents and their copies are made of different materials, resulting in different images of the documents when photographed. The document image can reflect certain surface information (such as brightness, content, and thickness) of the object being photographed (i.e., the genuine document or its copy). Based on this fact, this solution, after obtaining an image of the document to be tested, can detect the authenticity of the document based on the surface information obtained from the image.

It should be noted that in this solution, detecting the authenticity of the document to be tested means determining whether the document is a genuine document or a photocopy. A genuine document should be understood as a real card-type document corresponding to the target document. Whether the document to be tested matches the user's real identity (i.e., whether the document to be tested is indeed a genuine document held by the user) can be further determined through other methods, and this solution does not focus on that. The photocopy can be of the user's own genuine document, another person's genuine document, or even an image of any genuine document (such as a photocopy of a secondary image obtained from photographing a document image). This embodiment does not impose any limitations on this.

In this embodiment, the verification party can obtain the image of the document to be detected in response to an operation request. For example, when a user uses a client (running on a terminal device) to access a business platform (such as a financial platform, social platform, etc.) to register an account, modify account information, or perform operations such as transferring funds, the financial platform needs to verify the user's identity. In this case, the financial platform's server can initiate an identity verification request (i.e., an operation request) to the verification party, which can be used to trigger the verification party to verify the user's identity. Accordingly, if the verification party is a detection server, the detection server can, in response to the received identity verification request, instruct the client to obtain the image of the document to be detected; if the verification party is a client, the client can, in response to the received identity verification request, obtain the image of the document to be detected itself.

Of course, since users often possess multiple identification documents, the aforementioned identity verification request can specify which document (hereinafter referred to as the target document) needs to be detected. The client can then output the target document information (such as the document required for verification by a financial platform) to the user, including the document name, so that the user knows which document to use for verification. For example, the client can display the message "Please present your second-generation resident ID card" to inform the user that they need to use their second-generation resident ID card for further verification. Another example is displaying a sample photo of a social security card on the client, so the user knows they need to use their social security card for further verification. Furthermore, the client can also output the target document information through voice prompts, such as playing a message like "Please take a picture of your second-generation resident ID card," which will not be elaborated further.

The aforementioned identity verification request can be associated with a first preset operation. For example, in the case of a user registering an account, the request can be associated with a new account registration operation, such as redirecting to the account information entry page, displaying the platform's registration agreement for new accounts, or redirecting to the new account login page. As another example, in the case of a user modifying account information, the request can be associated with an account information modification operation, such as redirecting to an account information modification page, displaying the information modification rules, or displaying the modified account information. Yet another example, in the case of a user performing a transfer operation, the request can be associated with a transfer operation, such as prompting the user to specify a target account and transferring funds to that account. Accordingly, if the method described in this specification determines that the document to be detected is a genuine document, the verifier can execute the aforementioned first preset operation corresponding to the identity verification request. Alternatively, the identity verification request can also be associated with a second preset operation, such as an alarm operation, a log backup operation, or a document image storage operation. Therefore, if it is determined that the document to be detected is not a genuine document (i.e., a photocopy), the verifier can refuse to execute the aforementioned first preset operation and instead execute the second preset operation to properly handle identity forgery incidents.

To ensure the authenticity of the acquired document images and the accuracy of subsequent detection results, the document images acquired in this solution need to be captured during the rotation of the document to be detected. For example, the client described in this embodiment can have image capture or video capture functions, so that the document image can be captured by the client. Specifically, the document to be detected can be used as the subject of the photograph, and the client can directly photograph the document to obtain an document image, or the client can photograph the document to obtain a rotating video, so that the verification party can extract video image frames from the rotating video as the document image. If the verification party is the client, the client can directly photograph the document image, or extract video image frames from the photographed rotating video as the document image. If the verification party is the detection server, the client can upload the photographed document image to the detection server, or upload the video image frames extracted from the photographed rotating video as the document image to the detection server.

When the verification party is the client, the client can capture an image of the identification document. For example, with the client's camera turned on and facing the document to be verified, the client's preview screen can display the document being scanned. The user can then hold and rotate the document, or place it on a rotating object and rotate it. Correspondingly, the client can capture an image of the document while its preview screen shows the document rotating. In this way, the client can acquire at least one image of the document during its rotation for subsequent verification.

During the process of capturing an ID document image, the user may rotate the document to be detected in various ways. For example, for a rectangular document, the user may rotate it around the line connecting the center points of the longer side, or around the line connecting the center points of the shorter side, or even rotate it in an irregular manner. To facilitate the subsequent acquisition of relevant document surface information based on the document image and improve the accuracy of document authenticity detection, a specific rotation method can be specified to the user. For example, the client can display rotation prompts in the display interface (such as the display interface corresponding to the aforementioned shooting preview screen) to explain the rotation method. These rotation prompts can take at least one form, such as text, GIF animation, animation, or live video, to simply and clearly inform the user how to rotate the document. The rotation prompts may include at least one of the following: rotation direction, rotation speed, and the distance range between the document and the client. Furthermore, if the actual rotation of the document under inspection does not match the rotation prompt message, the client can replace the prompt with another form (e.g., changing from text to animation) to remind the user again. This allows the user to rotate the document according to the prompt message, ensuring the client captures an image of the document at the preset angle or position. Alternatively, to avoid repeatedly disturbing the user, the rotation prompt message can be omitted, and the document image can be directly captured for subsequent processing.

The following example illustrates the rotation process of the document to be tested, using Figure 3 as an example. As shown in (1) of Figure 3, the terminal device 301 is positioned above the document to be tested 302. The rear camera of the terminal device 301 faces the front of the document to be tested 302, and the document image 304 is displayed in its preview screen 303. At this time, the terminal device 301 faces the document to be tested 302, and the document image 304 displayed in the preview screen 303 is a rectangular front of the document. It should be noted that the terminal device 301 and the document to be tested 302 shown in each figure in Figure 3 are right views viewed from the right side of the terminal device 301, that is, side views of the terminal device 301 and the document to be tested 302 viewed from the right side of the preview screen 303.

Let's assume that the rotation angle shown in (1) is 0°. Then, as the user's finger 305 rotates the document 302 to be tested in a clockwise direction (as shown by the arrow) with the line 306 connecting the midpoint of the short axis of the document 302 to be tested, the document images and corresponding right views corresponding to each rotation angle are shown in Figure 3 (2), (3), (4), (5), and (6), respectively. Among them, (3) corresponds to a rotation of 90°. At this time, the camera of the terminal device is facing the side of the document to be tested, and the corresponding document image is the side image of the document to be tested. (5) corresponds to a rotation of 180°. At this time, the camera of the terminal device is facing the back of the document to be tested, and the corresponding document image is the back of the rectangular document to be tested. When the rotation angle is (0°, 90°), (90°, 180°), and (180°, 270°), the corresponding document images are shown in Figure 3 (2), (4), and (6), respectively. Of course, the user can continue to rotate the document to be detected. For example, during a rotation of [0°, 360°] (i.e., one full rotation), the client can sequentially capture multiple images of the document. The above rotation process is usually continuous. If the client cannot extract the surface information of the document from any of the captured images (e.g., the rotation speed is too fast, resulting in an unclear image), the user can be instructed to rotate the document again to capture all the document images during the multiple rotations.

To facilitate the subsequent acquisition of document surface information based on the document image, the client can capture an image of the document to be inspected at a preset tilt angle. This preset tilt angle is the relative angle between the document and the client, and can be, for example, the angle between the plane containing the front or back of the document and the plane containing the preview image of the document (such as the screen).

To this end, the client can identify feature points of the document image displayed in the preview screen and determine the preview angle of the document based on the feature points. Then, if the identified preview angle is within a preset angle range, the client can capture the corresponding document image. For example, during the document rotation process, the client can detect feature points of the document image displayed in the preview screen in real time, such as the center point of the side of the document, the vertices of the four corners of the document and/or the center point of the document image, etc., and then identify the preview angle of the document to be detected based on the detected feature points and the lines connecting the feature points. If the preview angle is within a preset angle range, the client can capture the corresponding document image. If the preset angle range is [-5°, 5°], the client can use the captured document image within this range as the front image of the document to be detected, similar to (1) in Figure 3; while if the preset angle range is [87°, 93°], the client can use the captured document image within this range as the side image of the document to be detected, similar to (3) in Figure 3. In this way, the client can flexibly capture the image of the document corresponding to the image to be detected within a preset angle range during the document rotation process.

During the rotation of the document to be inspected, the document and/or the client often shake. Therefore, to ensure the accuracy of the shooting angle, the client can display a positioning marker in the shooting preview screen. The user can then adjust the posture of the document according to this positioning marker, ensuring that the relative angle between the document and the client meets the aforementioned preset tilt angle. As the user adjusts the posture of the document, the document preview position in the shooting preview screen (i.e., the position of the document image within the shooting preview screen) will change accordingly. Thus, the client can capture an image of the document when its preview position matches the positioning marker. The aforementioned positioning marker corresponds to the aforementioned preset tilt angle.

As shown in Figure 3, the positioning marks are as shown in (1) and (3) respectively when the preset tilt angle is 0° and 90°. Taking (1) as an example, the four positioning marks 307 shown in (1) form a rectangle, and the aspect ratio of this rectangle is the same as that of the target document. When the preset tilt angle is 180°, the positioning marks shown in (5) are similar to those in (1), and will not be described again. When the preset tilt angle is 30°, the positioning marks are as shown in (2) with four positioning marks 308. Since the mobile phone camera is point-shaped relative to the sheet-like document, the camera achieves a point-shaped perspective during the shooting process. Therefore, the four positioning marks 308 form an isosceles trapezoid, and the angle between the plane of the isosceles trapezoid and the plane of the shooting preview screen of the document to be tested is 30°.

Regarding the positioning markers displayed in the preview screen, users can adjust their rotation angle while rotating the ID card to align the ID card image with the rectangle or isosceles trapezoid formed by the positioning markers. At this point, the ID card preview position matches the corresponding positioning marker. The client can capture an image of the ID card when it detects that the ID card preview position matches the corresponding positioning marker. The angle between the ID card and the terminal device at the time of capture is the preset angle corresponding to the positioning marker. For example, if the ID card image is located within the rectangle formed by the positioning markers 307, the angle between the ID card and the terminal device at the time of capture is 0°; as another example, if the ID card image is located within the isosceles trapezoid formed by the positioning markers 308, the angle between the ID card and the terminal device at the time of capture is 30°. The preset angle represented by the positioning markers can be set according to the ID card image; this manual does not limit its display position or specific preset angle.

Of course, when the verification party is the client, the client can also capture a rotation video of the document to be detected during its rotation process, and select the document image from the video frame of the rotation video. Because the rotation video records the rotation process of the document to be detected, the document image in some video frame also meets the above-mentioned preset tilt angle, document position, and other conditions. Therefore, the client can select the video frame that meets the above conditions as the document image by detecting the document image in the video frame. The specific process is not essentially different from the aforementioned embodiment, and will not be described again.

Step 204: Obtain the surface information of the document to be detected based on the document image.

Step 206: If the surface information of the document meets the preset detection conditions, determine that the document to be detected is a genuine document.

In this embodiment, after obtaining the image of the document to be tested, the verifier can further obtain the surface information of the document based on the image, and then determine the authenticity of the document by judging whether the surface information meets preset detection conditions. Specifically, if the surface information meets the preset detection conditions, the document to be tested can be determined to be a genuine document; if the surface information does not meet the preset detection conditions, the document to be tested can be determined to be a counterfeit copy.

The surface information of the document to be tested can take various forms, such as document thickness, document brightness, or document content. The corresponding preset testing conditions also differ for different forms of document surface information. The verifier can use at least one of the above forms of document surface information to determine whether the document to be tested is genuine. The process of determining the authenticity of the document to be tested using different forms of document surface information is explained below.

A. The information on the surface of the document is the document thickness.

Genuine identification documents are typically photocopied onto paper. Even if the photocopy is a high-resolution color copy, or even if the copied image is a 1:1 scale image of the genuine document, or if the photocopy has been cropped to match the edges of the original image, the thickness of the paper used for photocopying is usually much less than the thickness of the genuine document. Based on this fact, the verifier can use the document's thickness to determine its authenticity. For example, the verifier can use the document's thickness as surface information, with a preset detection condition that the document thickness is not less than a preset thickness threshold.

For example, the verifier can first determine the side image of the document to be tested from the aforementioned document image, wherein the side image contains the document image corresponding to the side view of the document to be tested (i.e., the image corresponding to the side view of the document to be tested). Of course, during the process of obtaining the aforementioned document image, the verifier can obtain an image containing the document image corresponding to the side view of the document to be tested, as the side image of the document to be tested. Taking the tilt angle of the document to be tested as 90° as an example, the corresponding side image is shown in Figure 3 (3), where document image 309 is the side view of the document to be tested at the current moment. Of course, the tilt angle of the document to be tested can also be 270°, or it can be within a preset angle range adjacent to 90° or 270° (such as [80°, 100°], or [265°, 275°], etc.). This embodiment of the specification does not limit this, as long as the aforementioned side image contains the document image corresponding to the side view of the document to be tested.

Furthermore, the verifier can determine the document thickness in various ways. As an exemplary embodiment, the verifier can determine the length and width of the side of the document to be tested based on the side image, and use the ratio of width to length as the document thickness. Specifically, the verifier can use the product of the number of pixels corresponding to the long and short sides of the side image and the width of a single pixel as the length and width, respectively. Of course, since the document thickness is the ratio of width to length, the ratio of the number of pixels corresponding to the long and short sides of the side image can also be directly used as the document thickness, thereby speeding up the determination of the document thickness. Correspondingly, the thickness threshold can be preset based on the standard document size of the target document. The verifier can determine that the document to be tested is a genuine document if the determined document thickness is not less than this thickness threshold. Taking the standard document size of the target document as 80mm × 50mm × 1mm as an example, considering that 1/80 = 1.25% < 1/50 = 2%, the verifier can preset the thickness threshold to 1.25%. Therefore, if the thickness of the document is detected to be not less than the thickness threshold, the verifier can determine that the document under test is genuine; otherwise, if the thickness of the document is detected to be less than the thickness threshold, the verifier can determine that the document under test is a forged copy. Alternatively, the verifier can preset a first thickness threshold of 1.25% (or 1.2%, etc.) for the long side of the document and a second thickness threshold of 2% (or 1.99%, etc.) for the short side. Then, if the thickness of the document under test corresponding to the long side is detected to be not less than the first thickness threshold, or if the thickness of the document under test corresponding to the short side is detected to be not less than the second thickness threshold, the verifier can determine that the document under test is genuine. Conversely, if the thickness of the document under test is less than both the first and second thickness thresholds, the verifier can determine that the document under test is a forged copy.

As another exemplary embodiment, the verifier can determine the width and current rotation angle of the side of the document to be tested based on the aforementioned side image, and determine the actual thickness value of the document to be tested based on the determined width and current rotation angle, as the document thickness. Accordingly, the aforementioned thickness threshold can be preset according to the standard document size of the target document. The verifier can determine that the document to be tested is a genuine document if the determined document thickness is not less than the thickness threshold. Taking the standard document size of the target document as 80mm × 50mm × 1mm as an example, the thickness threshold can be preset to 1mm (or 0.95mm, 0.8mm, etc.). If the detected actual thickness value is not less than the thickness threshold, the verifier can determine that the document to be tested is a genuine document; otherwise, if the detected document thickness is less than the thickness threshold, the verifier can determine that the document to be tested is a counterfeit copy.

As can be seen from the above embodiments, the verification party can determine the thickness of the document to be tested using only a single side image. Of course, to minimize errors, the verification party can also obtain multiple side images to determine multiple independent document thicknesses, and use the average of the multiple independent document thicknesses as the final detection result of the document thickness to be tested, which will not be elaborated further.

In addition, the verifier can acquire multiple images of the document under inspection at various rotation angles, and based on these images, obtain surface information such as the document's thickness and/or brightness. This surface information can then be used to detect the authenticity of the document. These details are explained below:

B. The information on the surface of the document is the brightness of the document.

Genuine identification documents are typically made of materials such as PVC (Polyvinyl Chloride), ABS (Acrylonitrile Butadiene Styrene), and PET (Polyethylene Terephthalate). Furthermore, the surface of these documents usually features anti-counterfeiting marks such as highly reflective laser patterns. When printing images of genuine documents, they are typically printed on paper; therefore, counterfeit copies are usually made of paper. Generally, compared to copies printed on rougher paper, genuine documents made of smooth materials like PVC have a higher reflectivity. Therefore, when the document being tested is rotated, the brightness change of a genuine document is often greater than that of a copy. For example, the brightness change of the laser patterns on a genuine document is significant, while the overall brightness change in a copy is not obvious. Based on this fact, the verifier can determine the authenticity of the document by its brightness. For example, the verifier can use the brightness of the document to be tested as the surface information of the document. The corresponding preset detection condition can be that the brightness variation among multiple document brightness values determined based on multiple document images is not less than a preset variation threshold.

The verifier can determine the aforementioned brightness change based on the brightness of multiple document images. For example, the verifier can determine the brightness of the document under test at corresponding rotation angles based on the multiple images, and then calculate the brightness difference between adjacent rotation angles as the brightness change. The angle difference between the rotation angles corresponding to the multiple document images can be set relatively small. For example, during the rotation of the document under test, multiple images can be continuously captured in a short period, such as every 0.5s or 0.2s, to avoid brightness changes caused by excessive angle changes masking brightness changes caused by material factors. Furthermore, the pixel color values of all the document images belong to the same color space, such as RGB (Red-Green-Blue) values or grayscale values for each pixel. The verifier can convert all RGB images to grayscale images to calculate the document brightness using the grayscale values of the pixels.

Taking two grayscale-processed ID card images as an example, the verifier can first determine the first image region corresponding to the ID card image in the first image and the second image region corresponding to the ID card image in the second image. Then, the verifier calculates the first grayscale mean of each pixel in the first image region and the second grayscale mean of each pixel in the second image region. The grayscale difference between the first and second grayscale mean values can then be used as the brightness change between the two ID card images. Accordingly, if this brightness change is not less than a preset threshold, the verifier can determine that the ID card to be detected is genuine.

Of course, the aforementioned brightness change can also be calculated using color values in other color spaces. For example, for any RGB image, the average R, average G, and average B values of each pixel in the corresponding image area can be calculated separately, and the average of the three can be used as the brightness value of the image. The difference between this brightness value and the brightness values of other RGB images can be used as the brightness change between the two images. Alternatively, multiple brightness changes can be calculated from three or more document images, and if each brightness change is not less than a preset change threshold, the document to be detected is determined to be a genuine document. Further details are omitted here.

C. The information on the surface of the document is the content of the document.

For the target document specified in the authentication request, its content (such as the text and photo printed on the document) is usually fixed. Furthermore, for the same target document, the content on the front and back matches each other. Similarly, the content on the front and back of a genuine document also matches each other. However, when a genuine document is printed on paper, usually only one side of the document (either the front or back) is printed on one side of the paper, or both sides are printed on the same side of the paper—the other side of the paper containing the photocopy is blank or contains other content—in this case, the content on the two sides of the photocopy does not match. Based on this fact, the verifier can detect the authenticity of the document by examining its content. For example, the verifier can use the brightness of the document as surface information, and the document content includes a first content corresponding to one side of the document and a second content corresponding to the other side. The corresponding preset detection condition can be that the first content matches the second content.

The verification party can pre-obtain a set of standard document content, which includes multiple interrelated standard contents corresponding to at least one type of document. Any of the standard contents may include text and/or images arranged in a preset position. As shown in Figure 4, the front of the document contains three front standard contents: an icon 401 in the upper left area, text 402 in the upper right area, and text 403 in the lower middle area. The back of the document contains three back standard contents: text 404 in the upper area, a photo 405 in the upper area, and text 406 in the lower area. The aforementioned front and back standard contents represent the content contained on the front and back of the same document, respectively, and are interrelated. The set of standard document content can record the shape of the document image containing each document content (a rectangle as shown in Figure 4), the position coordinates of each document content within the document image, and the type of each document content (image or text), etc. This standard document content set can record the standard content of various documents such as ID cards, social security cards, and public transport cards. Multiple standard content sections of any document (such as the front and back standard content mentioned above) are interconnected. Understandably, this set can record the standard content corresponding to as many documents as possible to broaden the applicability of the document authenticity verification system.

After obtaining a first image of the document containing either side (front or surface) of the document to be tested and a second image of the document containing the other side of the document to be tested, the verifying party can extract first content and second content from the first and second images, respectively. It is understood that if either side is the front of the document to be tested, the other side is the back; or, if either side is the back, the other side is the front. For example, a first image corresponding to the front of the document to be tested can be taken when the tilt angle of the document to be tested is 0°, and a second image corresponding to the back of the document to be tested can be taken when the tilt angle of the document to be tested is 180°. Content can be extracted from either image using OCR (Optical Character Recognition) technology, the specific process of which will not be elaborated here.

Furthermore, after extracting the first and second contents of the document to be tested, the verifier can query the set of standard document contents for the first standard content that matches the first content. If no first standard content is found (indicating that the set of standard document contents does not record the standard content of the standard document corresponding to the document to be tested), the document content processing for the document to be tested can be terminated directly, and the authenticity of the document to be tested can be determined through other methods. If the first standard content is found, the second standard content associated with the first standard content can be determined, and then it can be further determined whether the second content matches the second standard content: if the second content matches the second standard content, then it can be determined that the first content matches the second content.

Taking Figure 4 as an example, after extracting the first and second contents of the document to be tested, the standard content matching the first content can be queried in the standard document content set. If the first content of the document to be tested matches the front standard content shown in Figure 4, the back standard content associated with the front standard content can be determined; and further, it can be determined whether the second content of the document to be tested matches the back standard content: if they match, the document to be tested is determined to be a genuine document; otherwise, if they do not match, the document to be tested is determined to be a forged photocopy. Alternatively, if the first content of the document to be tested matches the back standard content shown in Figure 4, the front standard content associated with the back standard content can be determined; and further, it can be determined whether the second content of the document to be tested matches the front standard content: if they match, the document to be tested is determined to be a genuine document; otherwise, if they do not match, the document to be tested is determined to be a forged photocopy.

This concludes the description of the process for detecting the authenticity of a document by using only its thickness, brightness, and content as surface information. It is understood that the surface information may also include multiple surface information elements such as thickness, brightness, and content. In this case, the verifier can extract each surface information element from the document image and use each thickness element individually to detect the authenticity of the document, obtaining independent detection results. These independent results are then used to jointly determine the authenticity of the document, and this determination is taken as the final detection result. The specific process of using any one surface information element individually to detect the authenticity of the document can be found in sections A, B, and C above, and will not be repeated here.

For example, when considering multiple surface information elements—including document thickness, brightness, and content—as surface information, if the independent test results for each of these surface elements all indicate that the document is genuine, then the document can be determined to be genuine. Conversely, if any independent test result for any surface element indicates that the document is a counterfeit copy, then the document can be determined to be a counterfeit copy. Alternatively, if the independent test results are expressed as probability values, the accuracy of the independent test results for document thickness, brightness, and content can be statistically analyzed based on historical test records, and a weight value positively correlated with the accuracy can be assigned to each independent test result. The verifying party can then calculate a weighted average of the independent test results based on these weight values to determine the authenticity of the document.

Through the above embodiments, the verifier can obtain an image of the document under test captured during the rotation of the document, and obtain the document surface information based on this image. If the document surface information meets preset detection conditions, the verifier can determine that the document under test is genuine. It is understood that because genuine documents differ significantly from photocopies in material, size, etc., and these differences lead to differences in their surface information, this solution, after obtaining the document image and its surface information, can accurately distinguish whether the document under test is genuine based on this information, exhibiting high detection accuracy and effectively identifying counterfeit document attacks using photocopies.

Figure 5 is a schematic structural diagram of a device provided in an exemplary embodiment. Referring to Figure 5, at the hardware level, the device includes a processor 502, an internal bus 504, a network interface 506, a memory 508, and a non-volatile memory 510, and may also include other hardware required for services. One or more embodiments of this specification can be implemented in software, for example, the processor 502 reads the corresponding computer program from the non-volatile memory 510 into the memory 508 and then runs it. Of course, in addition to software implementation, one or more embodiments of this specification do not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. That is to say, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

Please refer to Figure 6. In one software implementation, the device for detecting the authenticity of documents may include:

Image acquisition unit 601 is used to acquire an image of the document to be inspected, the image of which is acquired during the rotation of the document to be inspected;

The information acquisition unit 602 is used to acquire the surface information of the document to be detected based on the document image;

The authenticity determination unit 603 is used to determine that the document to be tested is an authentic document when the surface information of the document meets the preset detection conditions.

Optionally, the image acquisition unit 601 is further configured to:

Acquire multiple images of the document to be detected at various rotation angles.

Optionally, the document surface information includes document brightness, and the preset detection condition includes: the brightness variation among the multiple document brightness values determined based on the multiple document images is not less than a preset variation threshold.

Optionally, the information acquisition unit 602 is further configured to:

The brightness of the document to be detected at the corresponding rotation angle is determined based on the multiple images.

Calculate the brightness difference between the document brightness at adjacent rotation angles, and use it as the brightness change.

Optionally, the document surface information includes document content, which includes first content corresponding to one side of the document to be tested and second content corresponding to the other side; the preset detection conditions include: the first content matches the second content.

Optionally, the real determination unit 603 is further configured to:

The standard document content set is used to query a first standard content that matches the first content and a second standard content that is associated with the first standard content. The standard document content set includes multiple interrelated standard contents corresponding to at least one type of document.

If the second content matches the second standard content, then the first content is determined to match the second content.

Optionally, any standard content includes: text and/or images arranged in preset positions.

Optionally, the document surface information includes the document thickness; the preset detection conditions include: the document thickness is not less than a preset thickness threshold.

Optionally, the information acquisition unit 602 is further configured to:

Determine the side view in the document image;

The length and width of the side of the document to be inspected are determined based on the side image, and the ratio of the width to the length is taken as the thickness of the document; or,

The width and current rotation angle of the side of the document to be inspected are determined based on the side image, and the actual thickness value of the document to be inspected is determined based on the width and the current rotation angle, which is used as the thickness of the document.

Optionally, the image acquisition unit 601 is further configured to:

If the document to be inspected is shown rotating in the preview screen, take an image of the document.

Optionally, it also includes:

The information display unit 604 is used to display rotation prompt information, which describes the rotation method of the document to be inspected.

Optionally, the image acquisition unit 601 is further configured to:

A location marker is displayed in the shooting preview screen, and the document image is captured when the document preview position matches the location marker.

Optionally, the image acquisition unit 601 is further configured to:

Identify the feature points of the ID card image displayed in the shooting preview screen, and determine the preview angle of the ID card based on the feature points;

When the preview angle is within a preset angle range, the image of the document is captured.

Optionally, the image acquisition unit 601 is further configured to:

Obtain the rotation video corresponding to the rotation process of the document to be detected, and select the document image from the video image frames of the rotation video.

Optionally, it also includes:

The forgery determination unit 605 is used to determine that the document to be tested is a forged photocopy when the surface information of the document does not meet the preset detection conditions.

Optionally, the image of the document to be detected is acquired in response to an operation request, and the device further includes:

The operation execution unit 606 is used to perform a preset operation corresponding to the operation request when it is determined that the document to be detected is a genuine document.

The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer, which can take the form of a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email sending and receiving device, game console, tablet computer, wearable device, or any combination of these devices.

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

Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and/or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

Computer-readable media, including both permanent and non-permanent, removable and non-removable media, can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, 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 technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage, quantum memory, graphene-based storage media or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

It should also be noted that 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 limitation, 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.

The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

The terminology used in one or more embodiments of this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of one or more embodiments of this specification. The singular forms “a,” “described,” and “the” used in one or more embodiments of this specification and in the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and/or” as used herein refers to and includes any or all possible combinations of one or more associated listed items.

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

The above description is merely a preferred embodiment of one or more embodiments of this specification and is not intended to limit the scope of one or more embodiments of this specification. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments of this specification should be included within the scope of protection of one or more embodiments of this specification.

Claims (16)

1. A method for verifying the authenticity of a document, comprising: Acquire an image of the document to be inspected, the image being captured during the rotation of the document to be inspected; The step of acquiring the document image of the document to be detected includes: acquiring multiple document images of the document to be detected at various rotation angles; Based on the document image, obtain the document surface information of the document to be detected, including the document brightness; If the surface information of the document meets the preset detection conditions, the document to be detected is determined to be a genuine document. The preset detection conditions include: the brightness variation among the brightness of multiple documents determined according to the multiple document images is not less than a preset variation threshold. Determining the brightness change includes: determining the brightness of the document to be detected at corresponding rotation angles based on the multiple document images, and calculating the brightness difference between adjacent rotation angles as the brightness change. 2. The method according to claim 1, wherein the document surface information includes document content, the document content includes first content corresponding to one side of the document to be detected and second content corresponding to the other side; the preset detection condition includes: the first content matches the second content. 3. The method according to claim 2, wherein determining that the first content matches the second content is accomplished by means of: The standard document content set is used to query a first standard content that matches the first content and a second standard content that is associated with the first standard content. The standard document content set includes multiple interrelated standard contents corresponding to at least one type of document. If the second content matches the second standard content, then the first content is determined to match the second content. 4. The method according to claim 2, wherein any standard content includes: text and/or images arranged in a preset position. 5. The method according to claim 1, wherein the document surface information includes document thickness; the preset detection condition includes: the document thickness is not less than a preset thickness threshold. 6. The method according to claim 5, wherein determining the thickness of the document comprises: Determine the side view in the document image; The length and width of the side of the document to be inspected are determined based on the side image, and the ratio of the width to the length is taken as the thickness of the document; or, The width and current rotation angle of the side of the document to be inspected are determined based on the side image, and the actual thickness value of the document to be inspected is determined based on the width and the current rotation angle, which is used as the thickness of the document. 7. The method according to claim 1, wherein acquiring the document image of the document to be detected includes: If the document to be inspected is shown rotating in the preview screen, take an image of the document. 8. The method according to claim 7, further comprising: Display rotation prompt information, which is used to describe the rotation method of the document to be tested. 9. The method according to claim 7, wherein capturing the image of the document comprises: A location marker is displayed in the shooting preview screen, and the document image is captured when the document preview position matches the location marker. 10. The method according to claim 7, wherein capturing the image of the document comprises: Identify the feature points of the ID card image displayed in the shooting preview screen, and determine the preview angle of the ID card based on the feature points; When the preview angle is within a preset angle range, the image of the document is captured. 11. The method according to claim 1, wherein acquiring the document image of the document to be detected includes: Obtain the rotation video corresponding to the rotation process of the document to be detected, and select the document image from the video image frames of the rotation video. 12. The method according to claim 1, further comprising: If the information on the surface of the document does not meet the preset detection conditions, the document to be tested is determined to be a counterfeit photocopy. 13. The method according to claim 1, wherein the image of the document to be detected is acquired in response to an operation request, the method further comprising: If the document to be tested is determined to be a genuine document, a preset operation corresponding to the operation request is executed. 14. A device for detecting the authenticity of a document, comprising: The image acquisition unit is used to acquire the document image of the document to be detected. The document image is acquired during the rotation of the document to be detected. Specifically, it is used to acquire multiple document images of the document to be detected at various rotation angles. An information acquisition unit is used to acquire surface information of the document to be detected based on the document image, wherein the surface information includes the document brightness; The authenticity determination unit is used to determine that the document to be detected is an authentic document when the surface information of the document meets the preset detection conditions. The preset detection conditions include: the brightness variation among the brightness of multiple documents determined according to the multiple document images is not less than a preset variation threshold. The information acquisition unit is further configured to: determine the brightness of the document to be detected at corresponding rotation angles based on multiple images, and calculate the brightness difference between adjacent rotation angles as the brightness change. 15. An electronic device comprising: processor; Memory used to store processor-executable instructions; The processor implements the method as described in any one of claims 1-13 by executing the executable instructions. 16. A computer-readable storage medium having stored thereon computer instructions that, when executed by a processor, implement the steps of the method as claimed in any one of claims 1-13.