CN112434727A

CN112434727A - Identity document authentication method and system

Info

Publication number: CN112434727A
Application number: CN202011208717.7A
Authority: CN
Inventors: 黄江波
Original assignee: Alipay Labs Singapore Pte Ltd
Current assignee: Ant Shield Co ltd
Priority date: 2020-05-02
Filing date: 2020-11-03
Publication date: 2021-03-02
Also published as: SG10202004041SA

Abstract

The application provides an Identification (ID) document authentication method and system. The method comprises the following steps: aligning the first identification document image and the second identification document image; cropping a first region of interest from the aligned first authentication document image, wherein the first region of interest includes a multi-angle security feature; performing a search to locate a second region of interest on the aligned second identification document image, wherein the second region of interest corresponds to the first region of interest; calculating a similarity score between the first region of interest and the second region of interest; and determining the authenticity of the identification document based on the comparison of the similarity score to the similarity threshold.

Description

Identity document authentication method and system

Technical Field

This document relates broadly, but not exclusively, to an identification document authentication method and an identification document authentication system.

Background

Electronically aware your customer (eKYC) is a digital due diligence process performed by a business entity or service provider to verify the identity of its customers to prevent identity fraud. The eKYC process typically includes a series of checks to verify its identity at an initial stage of establishing a relationship with the customer. Many eKYC processes involve potential customers submitting photographs of their official Identification (ID) documents, e.g., identification cards, driver's licenses, passports, and the like. The photograph may then be analyzed to verify the identity of the customer.

In a typical eKYC process, a customer is required to take a picture of his identification document. However, some attackers may use fraudulent identification documents or high resolution copies of authentic identification documents in the eKYC process.

Accordingly, there is a need for improving the manner in which identification documents can be authenticated.

Disclosure of Invention

Embodiments seek to provide an Identification (ID) document authentication method and an identification document authentication system that involve verifying multi-angle security features of an identification document to detect attacks using a replica of the identification document. A copy of the identification document (copy) may comprise a high resolution copy of the authentic identification document (copy) or a counterfeit identification document.

According to one embodiment, there is provided an Identification (ID) document authentication method, including: aligning the first identification document image and the second identification document image; cropping a first region of interest from the aligned first authentication document image, wherein the first region of interest includes a multi-angle security feature; performing a search to locate a second region of interest on the aligned second identification document image, wherein the second region of interest corresponds to the first region of interest; calculating a similarity score between the first region of interest and the second region of interest; and determining the authenticity of the identification document based on a comparison between the similarity score and a similarity threshold.

According to another embodiment, there is provided an Identification (ID) document authentication system including: an alignment device configured to align the first identification document image and the second identification document image; an image cropping device configured to crop a first region of interest from the aligned first credential image, wherein the first region of interest includes a multi-angle security feature; a search device configured to perform a search to locate a second region of interest on the aligned second identification document image, wherein the second region of interest corresponds to the first region of interest; a score calculation device configured to calculate a similarity score between the first region of interest and the second region of interest; and an authentication device configured to determine authenticity of the identification document based on a comparison between the similarity score and a similarity threshold.

Drawings

The embodiments are provided by way of example only and will be better understood and readily apparent to those of ordinary skill in the art from the following written description when read in conjunction with the accompanying drawings, wherein:

fig. 1 is a flowchart illustrating an example of an Identification (ID) document authentication method according to an embodiment.

FIG. 2 is a schematic diagram illustrating an example identification document authentication method of FIG. 1.

FIG. 3 is a schematic diagram illustrating an example of an identification document authentication system according to an embodiment.

FIG. 4 shows a schematic diagram of a computer system suitable for use in performing at least some of the steps of the identification document authentication method.

Detailed Description

Embodiments will now be described, by way of example only, with reference to the accompanying drawings. Like reference numbers and designations in the drawings indicate like elements or equivalents.

Some portions of the description that follows are presented explicitly or implicitly in terms of algorithms and functional or symbolic representations of operations on data within a computer memory. These algorithmic descriptions and functional or symbolic representations are the means used by those skilled in the data processing arts to most effectively convey the substance of their work to others skilled in the art. An algorithm is here, and generally, considered to be a self-consistent sequence of steps leading to a desired result. These steps are those requiring physical manipulations of physical quantities such as electrical, magnetic or optical signals capable of being stored, transferred, combined, compared, and otherwise manipulated.

Unless specifically stated otherwise, and as will be apparent from the following, it is appreciated that throughout the present specification, discussions utilizing terms such as "receiving," "scanning," "computing," "determining," "replacing," "generating," "initializing," "outputting," or the like, refer to the action and processes of a computer system, or similar electronic device, that manipulates and transforms data represented as physical quantities within the computer system into other data similarly represented as physical quantities within the computer system or other information storage, transmission or display devices.

The present specification also discloses an apparatus for performing the operations of the method. Such apparatus may be specially constructed for the required purposes, or may comprise a computer or other device selectively activated or reconfigured by a computer program stored in the computer. The algorithms and displays (displays) presented herein are not inherently related to any particular computer or other apparatus. Various machines may be used with programs in accordance with the teachings herein. Alternatively, the construction of more specialized apparatus to perform the required method steps may be appropriate. The structure of a computer adapted to perform the various methods/processes described herein will appear from the description below.

Furthermore, the present specification also implicitly discloses a computer program, since it is obvious to a person skilled in the art that the individual steps of the methods described herein can be implemented by computer code. The computer program is not intended to be limited to any particular programming language and implementation thereof. It will be appreciated that a variety of programming languages and code therefor may be used to implement the teachings of the disclosure contained herein. Moreover, the computer program is not intended to be limited to any particular control flow. There are many other variations of computer programs that may use different control flows without departing from the spirit or scope of the present invention.

Furthermore, one or more steps of a computer program may be executed in parallel rather than sequentially. Such a computer program may be stored on any computer readable medium. The computer readable medium may include, for example, a magnetic or optical disk, a memory chip, or other storage device suitable for interfacing with a computer. The computer readable media may also include hardwired media such as those illustrated in the internet system, or wireless media such as those illustrated in the GSM mobile phone system. When the computer program is loaded and executed on such a computer, it effectively creates means for implementing the steps of the preferred method.

Electronically aware your customer (eKYC) is a digital due diligence process performed by a business entity or service provider to verify the identity of its customers to prevent identity fraud. Authentication can be considered a form of fraud detection in which the user's legitimacy is verified and a potential fraudster can be detected before fraud is carried out. Effective authentication can enhance the data security of the system, thereby protecting the digital data from unauthorized users.

In a typical eKYC process, a customer is required to take a picture of his Identification (ID) document (e.g., identification card, driver's license, passport, etc.). However, some attackers may use a replica of the identification document in the eKYC process. The copy of the identification document may comprise a high resolution copy of an authentic identification document or a counterfeit identification document.

In authentic identification documents, security features are often incorporated to reduce the risk of counterfeiting or other forms of fraud. Multi-angle security features such as holograms, multiple laser images, kineprints and optically variable inks are widely used as important security features in identification documents. The multi-angle security feature may display different colors and/or images when viewed from different angles. Some types of identification documents may have only one multi-angle security feature, while some types of identification documents may have more than one multi-angle security feature. The imitation of identification documents is generally not able to imitate the multi-angle security features of real identification documents. Thus, the multi-angle security feature of verifying identification documents may be used to detect attacks of counterfeits using identification documents.

Embodiments described herein may relate to an eKYC process that includes two general processes: a user registration process and a user authentication process. In the user registration process, the user may be required to take a picture of his identification document from different angles to submit as a user registration image. In the user authentication process, the same user may take a picture of his face (i.e., "self-portrait") and submit as a user authentication image.

According to one embodiment, a user may be required to take at least two photographs of his identification document from different angles to submit as a user registration image. There are many methods available for detecting attacks that use identification document images from different identification documents of the same user or identification document images of the same type of identification document from different users. For example, the detection method may be a comparison of textual information in the identification document image, or a comparison of a user's portrait in the identification document image using facial recognition techniques. However, a primary concern herein is authenticating an identification document based on an image of the same identification document.

In order to reduce the success rate of attacks using imitations of identification documents, an identification document authentication method may be implemented that involves comparing the similarities between regions of multi-angle security features on images of identification documents. The method may include a set of instructions for performing a sliding window search operation to locate an area of the multi-angle security feature on a second image of the user-submitted identification document. The areas of the multi-angle security feature on the first and second images of the identification document may then be compared to determine whether the identification document is authentic.

As described above, the imitation of identification documents does not typically emulate the multi-angle security features of authentic identification documents. The existing anti-fraud method for the identification document cannot fully utilize the multi-angle security features of the identification document to detect the attack by using the imitation of the identification document. The techniques described herein produce one or more technical effects. Particularly, by calculating the similarity between the multi-angle security feature areas on the first image and the second image of the identity document, the identity document authentication method and system can reduce the success rate of attacks on eKYC processing. If it is determined that the multi-angle security feature regions on the first and second images of the identification document are similar, the identification document may be identified as a replica of the identification document.

In addition, the identification document authentication method and system can provide higher accuracy in detecting attacks using replicas of the identification document. The similarity score is expected to be relatively high if a replica of the identification document without the multi-angle security feature is used. In some cases, the similarity score may be as high as 1. Thus, counterfeit identification documents can be accurately identified using identification document authentication methods and systems.

The identification document authentication method and system may also provide greater efficiency in detecting attacks using replicas of identification documents. The model may be trained to detect attacks using replicas of the identification document. However, in practice, collecting a relatively large number of identification document images at different angles can be time consuming and difficult. For each multi-angle security feature of the identification document, it can also be very time consuming and difficult to collect a relatively large amount of label training data in order to train the model for accuracy in detecting attacks. Unlike training models, the identification document authentication methods and systems provided herein may not require a large amount of training data. A threshold for determining similarity between multi-angle security feature regions on a first image and a second image of an identification document may be set using a relatively small amount of data. An optimal number of steps that can be achieved at a relatively high speed can also be used.

Fig. 1 is a flow diagram 100 illustrating an example of an Identification (ID) document authentication method according to an embodiment. At step 102, the first identification document image and the second identification document image are aligned. At step 104, a first region of interest is cropped from the aligned first credential image. The first region of interest includes a multi-angle security feature. At step 106, a search is performed to locate a second region of interest on the aligned second identification document image. The second region of interest corresponds to the first region of interest. In step 108, a similarity score between the first region of interest and the second region of interest is calculated. At step 110, the authenticity of the identification document is determined based on a comparison between the similarity score and a similarity threshold.

FIG. 2 is a diagram 200 illustrating an example identification document authentication method of FIG. 1. As described above, a user may be required to take at least two photographs of his identification document from different angles to submit as a user registration image. As shown in FIG. 2, the first identification document image 202 and the second identification document image 204 may be images of identification documents captured at different angles.

According to an embodiment, the step 102 of aligning the first identification document image 202 and the second identification document image 204 may comprise: an image regression (image regression) of the first identification document image 202 is performed to determine first coordinates of the four corners of the first identification document image 202, and an image regression of the second identification document image 204 is performed to determine second coordinates of the four corners of the second identification document image 204. The step 102 of aligning the first identification document image 202 and the second identification document image 204 may further comprise: image warping (image warping) of the first identification document image 202 is performed based on first coordinates of the four corner points, and image warping of the second identification document image 204 is performed based on second coordinates of the four corner points. For image regression, the identification document image may be used as an input, and the coordinates of the four corner points of the identification document image are possible outputs. For image warping, the identification document image and the coordinates of the four corner points of the identification document image may be used as inputs, and an aligned identification document image is a possible output. Furthermore, image warping may be performed using opencv warping.

The step 104 of cropping the first region of interest 208 from the aligned first authentication document image 206 may comprise: the method may include determining an identification document type based on a classification model, selecting an identification document template corresponding to the determined identification document type, defining a frame (frame) based on the identified identification document template, and cropping 208 a first region of interest from the aligned first identification document image 206 based on the defined frame.

A classification model for determining the type of identification document may be trained using a dataset of identification document images of different identification documents. The trained classification model may be able to determine the identification document type based on the identification document image transmitted thereto without having to extract features of the identification document and match the features to a data set. In some implementations, the trained classification model may be able to determine the identification document type based on the unaligned first identification document image 202 and the unaligned second identification document image 204 sent thereto. In other words, the identification document type may be determined prior to aligning the first identification document image 202 and the second identification document image 204.

In general, the same type of identification document may have the same template. There are many ways to obtain a template for an identification document type. For example, to obtain a template for a hong kong identification document, an image of the hong kong identification document may be captured or acquired from an available source. Referring to step 102 of FIG. 1, an image of a hong Kong identity document may be aligned to obtain a template of the hong Kong identity document.

Further, there may be many ways to define a border for cropping the first region of interest 208 from the aligned first identification document image 206 for a particular identification document type. For example, research may be conducted using available sources to determine the security features of hong kong identity documents, including multi-angle security features. Through research, the approximate location of the multi-angle security features on the hong Kong identity document can also be determined. The position coordinates of the multi-angle safety features on the hong Kong identity document can be marked by using the template of the hong Kong identity document obtained by the method. For example, the image size of the aligned identification document image may be 250 × 400 (height × width), and the position coordinates (x, y) of the multi-angle security feature may be (20, 135) in the upper left corner and (120, 210) in the lower right corner. These coordinates may define the bounding box of the cropped area of interest. For example, given a hong kong identification document image, after aligning the image to an image size of 250 x 400 (height x width), the defined bounding box may be used to crop the region of interest.

The step 106 of performing a search to locate the second region of interest 214 on the aligned second identification document image 210 may include: the size of the search window 212 is changed in a first set of steps. Each step of the first set of steps may comprise: the location of the search window 212 is changed in a second set of steps. A comparison score may be calculated at each step of the second set of steps, and the highest comparison score may be output as the similarity score. Calculating the comparison score may include: the first region of interest 208 is converted to a first grayscale image 216. The first grayscale image 216 may be represented by a first region of interest grayscale image matrix. Calculating the comparison score may further include: the second region of interest 214 is converted into a second gray scale image 218. The second grayscale image 218 may be represented by a second region of interest grayscale image matrix. Further, calculating the comparison score may include: performing normalization on the first region of interest grayscale image matrix and the second region of interest grayscale image matrix, and performing a dot product operation on the normalized first region of interest grayscale image matrix and the normalized second region of interest grayscale image matrix. Normalization and dot product operations can be a relatively fast method of measuring the distance between two images.

In some embodiments, the search to locate the second region of interest 214 on the aligned second identification document image 210 may be a sliding search. The second region of interest 214 located on the aligned second identification document image 210 by the search may be the same region of interest as the first region of interest 208 on the aligned first identification document image 206 and may generate a highest comparison score, which may be output as a similarity score. The same region of interest may refer to the same size first and second regions of

interest

208, 214 and the same location first and second regions of

interest

208, 214 on the aligned first and second

identification document images

206, 210, respectively. The comparison score may be calculated based on the first and second regions of

interest

208, 214 being the same size.

In some embodiments, the opencv function cv2.cvtcolor () may be used to convert the first region of interest 208 to a first grayscale image 216 and the second region of interest 214 to a second grayscale image 218. The first and second region of interest grayscale image matrices may be two-dimensional matrices. The normalization of the first region of interest grayscale image matrix may be performed by subtracting the average of the matrix from the matrix and then dividing the result by the standard deviation of the matrix. The normalization of the second region of interest grayscale image matrix may be performed using the same method.

As described above, the step 106 of performing a search to locate the second region of interest 214 on the aligned second identification document image 210 may include: the size of the search window 212 is changed in a first set of steps. The first set of steps for changing the size of the search window 212 may traverse (go through) a possible size ratio between the first region of interest 208 and the second region of interest 214. For identification document images captured from different angles, the size of the region of interest may not change. However, aligning the identification document images may result in a slight difference between the size of the first region of interest 208 and the size of the second region of interest 214. In other words, the size ratio between the first region of interest 208 and the second region of interest 214 may be approximately 1:1, but may not be exactly 1: 1. The first set of steps for changing the size of the search window 212 may traverse the size ratio range (0.9, 1.1), each step incremented by 0.01. In each step, the search window 212 may be resized by a size ratio. For example, after resizing, the size of the search window 212 may be h0 × w0 (height × width). In some embodiments, the size ratio range (0.9, 1.1) of the search window 212 may be fixed for a certain identification document type.

As described above, each of the first set of steps for changing the size of the search window 212 may include: the location of the search window 212 is changed in a second set of steps. The second set of steps may traverse possible locations of the second region of interest 214 on the aligned second identification document image 210. For example, referring to the bounding box defined for cropping the region of interest as described above, the size of the aligned second identification document image 210 may be 300 × 400 (height × width), and the defined bounding box may be approximately located at coordinates (y ═ 180, x ═ 40). A second set of steps for changing the position of the search window 212 may traverse the range of y coordinates (170, 190) and the range of x coordinates (30, 50), each step incremented by 1. At each step, the second region of interest 214 having the size of the resized search window 212 (h0 x w0, see the resized example of the search window 212 described above) may be cropped from the aligned second identification document image 210 at that location. In some embodiments, the search ranges for the y and x coordinates may be fixed for a certain identification document type.

As described above, a comparison score may be calculated at each step of the second set of steps, and the highest comparison score may be output as the similarity score. In some embodiments, the second set of steps may further include variables to store the highest comparison score calculated in the second set of steps. For example, the variable may be referred to as a MAX variable. In each of the second set of steps, the MAX variable may be updated with a higher comparison score if the calculated comparison score is higher than the MAX variable. On the other hand, if the calculated comparison score is below the MAX variable, the MAX variable may remain unchanged. After the second set of steps is completed, the MAX variable may be the highest comparison score and may be output as a similarity score.

If a large number of steps are provided for the first and second sets of steps, the overall process of the identification document authentication method may be relatively slow. To provide accuracy and speed, the first and second set of steps can be designed to be 20 steps each. In other words, the total number of steps may be 400.

According to an embodiment, the step 110 of determining the authenticity of the identification document may include authenticating that the identification document is authentic if the similarity score is below a similarity threshold.

The similarity threshold may be determined by the collected data. For example, authentic and counterfeit identification document samples may be collected. Using the identification document authentication methods provided herein, a similarity score for a genuine document and an identification document replica can be calculated to determine a similarity threshold. For example, if the similarity score of a genuine document is in the range of (0.5, 0.7) and the similarity score of an identification document replica is in the range of (0.7, 0.9), 0.7 may be set as a suitable similarity threshold.

In the case where the identification document to be authenticated is a real identification document having multi-angle security features, the region of interest may display different images when viewed from different angles. On the other hand, where the identification document to be authenticated is a counterfeit of the identification document (e.g., a high resolution copy of a genuine identification document or a counterfeit identification document), the counterfeit of the identification document may not have the multi-angle security feature. Therefore, the region of interest on the copy of the identification document shows the same image even when viewed from different angles, and the calculated similarity score can be as high as 1. As such, a similarity score below the similarity threshold may indicate that the first region of interest 208 is substantially different from the second region of interest 214, and thus the identification document may be authentic. Otherwise, the identification document may be a copy of the identification document.

In some embodiments, the graphics processing unit may be used in identification document authentication methods and systems. The first set of steps described above may remain the same, with each step providing a resized search window by a ratio. In a Convolutional Neural Network (CNN), the resized search window may be considered a kernel. Instead of using the second set of steps to crop the second region of interest, the possible crops of the second region of interest may be stacked into an H W C matrix. H and W may be the height and width of the second region of interest and C may be the number of possible crop. The H W C matrix can be viewed as a feature map of the CNN. A two-dimensional convolution can be performed using the obtained feature map and kernel. For such an embodiment, higher processing speeds of up to 400% can be achieved.

The term "configured to" is used herein in connection with systems, devices, and computer program components. For a system of one or more computers configured to perform a particular operation or action, it is meant that the system has installed thereon software, firmware, hardware, or a combination thereof that when executed causes the system to perform the operation or action. For one or more computer programs configured to perform particular operations or actions, it is meant that the one or more programs include instructions that, when executed by data processing apparatus, cause the apparatus to perform the operations or actions. For a specific logic circuit configured to perform a particular operation or action, it means that the circuit has electronic logic to perform the operation or action.

FIG. 3 is a diagram 300 that illustrates an example of an identification document authentication system, according to an embodiment. The document authentication system includes an alignment device 302 configured to align a first image of an identification document with a second image of the identification document. The identification document authentication system also includes an image cropping device 304, the image cropping device 304 configured to crop the first region of interest from the aligned first identification document image. The first region of interest includes a multi-angle security feature. The document authentication system also includes a search device 306, the search device 306 configured to perform a search to locate a second region of interest on the aligned second document image. The second region of interest corresponds to the first region of interest. Further, the identification document authentication system includes a score calculation device 308, the score calculation device 308 configured to calculate a similarity score between the first region of interest and the second region of interest. The identification document authentication system also includes an authentication device 310, the authentication device 310 configured to determine the authenticity of the identification document based on a comparison between the similarity score and a similarity threshold. The first identification document image and the second identification document image may be images of identification documents captured at different angles.

According to an embodiment, alignment apparatus 302 may be further configured to perform an image regression of the first authentication document image to determine first coordinates of four corner points of the first authentication document image. The alignment apparatus 302 may also be configured to perform an image regression of the second identification document image to determine second coordinates of four corner points of the second identification document image. Further, the alignment apparatus 302 may be configured to perform image warping of the first identification document image based on first coordinates of the four corner points and to perform image warping of the second identification document image based on second coordinates of the four corner points.

The image cropping device 304 may be further configured to determine an identification document type based on the classification model, select an identification document template corresponding to the determined identification document type, define a border based on the identified identification document template, and crop the first region of interest from the aligned first identification document image based on the defined border.

The search device 306 may also be configured to change the size of the search window in the first set of steps. Each step of the first set of steps may comprise: the position of the search window is changed in a second set of steps. The score calculation device 308 may also be configured to calculate a comparison score at each step of the second set of steps, and may output the highest comparison score as the similarity score. The score calculation device 308 may also be configured to convert the first region of interest into a first grayscale image. The first grayscale image may be represented by a first region of interest grayscale image matrix. The score computing device 308 may also be configured to convert the second region of interest into a second grayscale image. The second grayscale image may be represented by a second region of interest grayscale image matrix. Further, the score calculation device 308 may be configured to perform normalization on the first region of interest grayscale image matrix and the second region of interest grayscale image matrix. The score computation device 308 may also be configured to perform a dot product operation on the normalized first region of interest grayscale image matrix and the normalized second region of interest grayscale image matrix.

According to an embodiment, the authentication device 310 may be further configured to authenticate the identification document as authentic if the similarity score is below a similarity threshold.

FIG. 4 shows a schematic diagram of a computer system suitable for performing at least some of the steps of the identification document authentication method.

The following description of computer system/computing device 400 is provided by way of example only and is not intended to be limiting.

As shown in fig. 4, the exemplary computing device 400 includes a processor 404 for executing software routines. Although a single processor is shown for clarity, computing device 400 may also include a multi-processor system. The processor 404 is connected to a communication facility 406 for communicating with other components of the computing device 400. The communication facilities 406 may include, for example, a communication bus, cross-bar, or network.

Computing device 400 also includes a main memory 408, such as Random Access Memory (RAM), and a secondary memory 410. The secondary memory 410 may include, for example, a hard disk drive 412 and/or a removable storage drive 414, which may include a magnetic tape drive, an optical disk drive, etc. The removable storage drive 414 reads from and/or writes to a removable storage unit 418 in a well known manner. Removable storage unit 418 may comprise a magnetic tape, an optical disk, etc. which is read by and written to by removable storage drive 414. As will be appreciated by those skilled in the relevant art, the removable storage unit 418 includes a computer-readable storage medium having stored therein computer-executable program code instructions and/or data.

In alternative embodiments, secondary memory 410 may additionally or alternatively include other similar devices for allowing computer programs or other instructions to be loaded into computing device 400. Such devices may include, for example, a removable storage unit 422 and an interface 420. Examples of removable storage unit 422 and interface 420 include a removable memory chip (e.g., an EPROM, or PROM) and associated socket, and other removable storage units 422 and interfaces 420 that allow software and data to be transferred from removable storage unit 422 to computer system 400.

Computing device 400 also includes at least one communication interface 424. Communication interface 424 allows software and data to be transferred between computing device 400 and external devices via a communication path 426. In various embodiments, communication interface 424 allows data to be transferred between computing device 400 and a data communication network, such as a public or private data communication network. The communication interface 424 may be used to exchange data between different computing devices 400, which computing devices 400 form part of an interconnected computer network. Examples of communication interface 424 may include a modem, a network interface (e.g., an ethernet card), a communication port, an antenna with associated circuitry, and the like. The communication interface 424 may be wired or wireless. Software and data transferred via communications interface 424 are in the form of signals which may be electronic, electromagnetic, optical, or other signals capable of being received by communications interface 424. These signals are provided to the communications interface via communications path 426.

Optionally, computing device 400 also includes a display interface 402 that performs operations for presenting images to an associated display 430 and an audio interface 432 that performs operations for playing audio content via associated speakers 434.

As used herein, the term "computer program product" may refer, in part, to removable storage unit 418, removable storage unit 422, a hard disk installed in hard disk drive 412, or a carrier wave that carries software to communication interface 424 through communication path 426 (wireless link or cable). Computer-readable storage media refers to any non-transitory tangible storage medium that provides recorded instructions and/or data to computing device 400 for execution and/or processing. Examples of such storage media include floppy disks, magnetic tapes, CD-ROMs, DVDs, Blu-rays^TMA disk, a hard drive, a ROM or integrated circuit, USB memory, a magneto-optical disk, or a computer readable card such as a PCMCIA card, etc., whether or not such devices are internal or external to computing device 400. Examples of transitory or non-tangible computer-readable transmission media that may also participate in providing software, applications, instructions, and/or data to the computing device 400 include radio or infrared transmission channels and network connections to another computer or networked device, as well as the internet or intranet, including e-mail transmissions and information recorded on websites and the like.

Computer programs (also called computer program code) are stored in the main memory 408 and/or the secondary memory 410. Computer programs may also be received via communications interface 424. Such computer programs, when executed, enable computing device 400 to perform one or more features of embodiments discussed herein. In various embodiments, the computer programs, when executed, enable the processor 404 to perform the features of the embodiments described above. Accordingly, such computer programs represent controllers of the computer system 400.

The software may be stored in a computer program product and loaded into computing device 400 using removable storage drive 414, hard drive 412, or interface 420. Alternatively, the computer program product may be downloaded to computer system 400 over communications path 426. The software, when executed by the processor 404, causes the computing device 400 to perform the functions of the embodiments described herein.

It should be understood that the embodiment of fig. 4 is presented by way of example only. Thus, in some embodiments, one or more features of computing device 400 may be omitted. Furthermore, in some embodiments, one or more features of computing device 400 may be combined together. Additionally, in some embodiments, one or more features of computing device 400 may be separated into one or more component parts.

It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

Claims

1. An authentication method of an identification document, comprising:

aligning the first identification document image and the second identification document image;

cropping a first region of interest from the aligned first authentication document image, wherein the first region of interest includes a multi-angle security feature;

performing a search to locate a second region of interest on the aligned second identification document image, wherein the second region of interest corresponds to the first region of interest;

calculating a similarity score between the first region of interest and the second region of interest; and

determining authenticity of the identification document based on a comparison between the similarity score and a similarity threshold.

2. The method of claim 1, wherein the first identification document image and the second identification document image are images of an identification document captured at different angles.

3. The method of claim 1 or 2, wherein aligning the first identification document image and the second identification document image comprises:

performing an image regression of the first authentication document image to determine first coordinates of four corner points of the first authentication document image;

performing image regression of the second identification document image to determine second coordinates of four corner points of the second identification document image;

performing image warping of the first identification document image based on first coordinates of the four corner points; and

performing image warping of the second identification document image based on second coordinates of the four corner points.

4. The method of any of the above claims, wherein cropping the first region of interest from the aligned first identification document image comprises:

determining the type of the identification document based on the classification model;

selecting an identification document template corresponding to the determined identification document type;

defining a frame based on the identified identification document template; and

cropping the first region of interest from the aligned first credential image based on the defined bounding box.

5. The method of any of the above claims, wherein performing a search to locate the second region of interest on the aligned second identification document image comprises:

changing a size of a search window in a first set of steps, wherein each step in the first set of steps comprises changing a position of the search window in a second set of steps.

6. The method of claim 5, wherein,

calculating a comparison score in each step of said second set of steps, and

outputting the highest comparison score as the similarity score.

7. The method of claim 6, wherein calculating the comparison score comprises:

converting the first region of interest into a first grayscale image, wherein the first grayscale image is represented by a first region of interest grayscale image matrix;

converting the second region of interest into a second grayscale image, wherein the second grayscale image is represented by a second region of interest grayscale image matrix;

performing normalization on the first region of interest grayscale image matrix and the second region of interest grayscale image matrix; and

performing a dot product operation on the normalized first region of interest grayscale image matrix and the normalized second region of interest grayscale image matrix.

8. The method of any of the preceding claims, wherein determining the authenticity of the identification document comprises:

authenticating the identification document as authentic on condition that the similarity score is below the similarity threshold.

9. An identification document authentication system comprising:

an alignment device configured to align the first identification document image and the second identification document image;

an image cropping device configured to crop a first region of interest from the aligned first credential image, wherein the first region of interest includes a multi-angle security feature;

a search device configured to perform a search to locate a second region of interest on the aligned second identification document image, wherein the second region of interest corresponds to the first region of interest;

a score calculation device configured to calculate a similarity score between the first region of interest and the second region of interest; and

an authentication device configured to determine authenticity of the identification document based on a comparison between the similarity score and a similarity threshold.

10. The system of claim 9, wherein the first identification document image and the second identification document image are images of an identification document captured at different angles.

11. The system of claim 9 or 10, wherein the alignment device is further configured to:

12. The system of any of claims 9 to 11, wherein the image cropping device is further configured to:

defining a frame based on the identified identification document template; and

13. The system of any of claims 9 to 12, wherein the search device is further configured to:

the size of the search window is changed in a first set of steps, each step in the first set of steps comprising changing the position of the search window in a second set of steps.

14. The system of claim 13, wherein the score computing device is further configured to compute a comparison score in each step of the second set of steps and output a highest comparison score as the similarity score.

15. The system of claim 14, wherein the score computing device is further configured to:

16. The system of any of claims 9 to 15, wherein the authentication device is further configured to authenticate the identification document as authentic if the similarity score is below the similarity threshold.