Moire pattern detection in digital images and its biometric detection system

Abstract

The present invention relates to detection of Moiré patterns in digital images by processing multiple digital images of a captured object and / or scene, where each of the multiple digital images is taken at a different image exposure and / or resolution. Additionally, the present invention relates to methods and systems for assessing the liveness of a captured biometric feature based on detection of Moiré patterns from the multiple digital images.

Description

[Technical Field]

[0001] The present invention relates to the detection of moire patterns in digital images. More particularly, the present invention relates to the liveness detection of biometric features captured for biometric verification. [Background technology]

[0002] Moiré patterns are thought to be interference patterns caused by the overlapping of digital grids on a camera sensor, resulting in high-frequency noise in the image. Detecting and removing these patterns is a critical step in many applications. For example, detecting and removing moiré patterns is a key step in biometric verification systems, where these patterns can be used to determine the liveness of captured biometric features, thereby preventing spoofing attacks. Similarly, moiré patterns can be used in scanning applications to improve optical character recognition (OCR).

[0003] An example of a known method for detecting moiré patterns in digital images is presented in E. Abraham, "Moiré Pattern Detection using Wavelet Decomposition and Convolutional Neural Network," 2018 IEEE Symposium Series on Computational Intelligence (SSCI), Bangalore, India, 2018, pp. 1275-1279, doi:10.1109 / SSCI.2018.8628746. A disadvantage of current techniques for detecting moiré patterns is that their detection accuracy can be affected by the intensity profile of the digital image. This is because the frequency intensity of moiré patterns determines their visibility in a captured image and is highly dependent on the pixel intensity of the digital image. For example, the lower the pixel intensity of a captured image, and therefore the darker the captured image, the lower the frequency intensity of moiré patterns present in the image. A similar effect can be observed in the extent to which the captured image is overexposed. As a result, due to pixel intensity variations in digital images, Moiré patterns may not be accurately detected and / or extracted, which may be detrimental in preventing spoofing attacks in biometric verification systems. [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] E. Abraham, "Moire Pattern Detection using Wavelet Decomposition and Convolutional Neural Network", 2018 IEEE Symposium Series on Computational Intelligence (SSCI), Bangalore, India, 2018, pp. 1275-1279, doi:10.1109 / SSCI.2018.8628746 [Non-patent document 2] Raskar, Ramesh,"Computational Photography: Epsilon to Coded Photography",2009 Summary of the Invention [Problem to be solved by the invention]

[0005] It is an object of the present invention to provide a system and method for detecting moiré patterns in digital images that overcomes the disadvantages of prior art solutions.

[0006] It is a further object of the present invention to provide a system and method for verifying the liveness of a captured biometric feature of a person in a biometric verification system. [Means for solving the problem]

[0007] The object of the invention is achieved according to the invention by a system and a method which show the technical characteristics of the respective independent claims. Preferred embodiments of the invention are disclosed in the dependent claims.

[0008] According to one aspect of the present invention, a method for detecting moiré pattern information in a digital image is presented, the method comprising: receiving a set of digital images of an object and / or scene, the set of digital images including images captured by one or more cameras and at different resolutions and / or exposures; processing each digital image in the set to determine a portion of the imaged object and / or scene that contains Moiré pattern information at the corresponding image resolution and / or exposure; selecting a digital image from the set comprising a portion of an imaged object and / or scene that contains Moiré pattern information; and generating a merged digital image of the captured object and / or scene from the selected digital images, the merged digital image including the moiré pattern information detected in each of the selected digital images.

[0009] The method of the present invention enhances the detection of moiré patterns in digital images by processing and combining moiré pattern information detected from multiple digital images of the same object and / or scene captured at different image resolutions and / or exposures. Generally, pixel intensities of a captured object and / or scene vary within a captured digital image, resulting in portions of the image having different light / pixel intensity values. Therefore, regions of a digital image may be presented with different light / pixel intensity values. Moiré patterns depend on pixel intensities in a digital image. For example, moiré patterns may be difficult to detect in dark or overexposed regions of a captured image. Therefore, with previous techniques, only a small portion of a moiré pattern may be detected in a digital image due to camera sensor limitations and lighting conditions. Enhancing the detection of moiré patterns from regions of a digital image having different pixel intensity values ​​is important for a range of applications, including, but not limited to, biometric verification, optical character recognition (OCR), and the like. The method of the present invention overcomes the disadvantages of prior art solutions by extracting moiré pattern information from digital images of the same object and / or scene captured at different image resolutions and / or exposures. Therefore, it is possible to detect moiré patterns from different regions of a digital image. For example, by gradually increasing the image exposure between successive digital images, it is possible to increase the light intensity in darker areas of the digital images, which may reveal moiré patterns and lead to their accurate detection. Similarly, by varying the image resolution between successive digital images, it is possible to change the moiré patterns captured in the digital images. As such, different moiré patterns may appear depending on the image resolution, which may further enhance the detection of moiré patterns from the captured images and / or scenes. In the methods presented herein, moiré pattern information detected and / or extracted from a set of digital images is combined into a single digital image for a representation of the captured object and / or scene.A combined digital image may be generated by combining details from different captured digital images in a set that contain moiré patterns. For example, a composite digital image may be generated using high dynamic range imaging techniques known in the art, which involve capturing multiple images of the same scene using different exposure values ​​and then combining the images, including the moiré patterns, into a single image that represents the range of tonal values ​​within the captured scene and / or object. The composite digital image thus contains moiré pattern information from different ranges of digital images, thereby overcoming the disadvantages of prior art solutions.

[0010] According to embodiments of the present invention, detection of Moiré patterns may be performed using wavelet decomposition and / or multi-input deep convolutional neural networks (CNNs).

[0011] According to an embodiment of the present invention, processing the set of digital images includes aligning the objects and / or scenes captured in each of the digital images in the set. Typically, the digital images are aligned to compensate for any movement of the objects and / or scenes that occurred between successive digital images, such as movement of the heads of captured people. For example, alignment of the digital images may be performed using known image registration techniques known in the art, such as processing the digital images using a digital image alignment algorithm or another known technique. For example, image registration may be based on feature registration, pixel-based alignment, or any other known technique.

[0012] According to an embodiment of the present invention, processing the digital images includes dividing each digital image into a grid of predetermined dimensions. According to an embodiment of the present invention, processing the digital images includes detecting grid portions in each digital image that contain Moiré pattern information. The division facilitates the temporal analysis of the digital images in the set, thereby simplifying the detection of similar regions in the set of digital images. Thus, portions of the digital images that contain Moiré pattern information can be easily detected and compared with corresponding portions of the remaining digital images. It should be noted that image division can be performed in various ways. For example, the image analysis can be grid-based, where the digital image is divided into small portions, or it can be object-division-based, where the digital image is analyzed to detect similar regions, such as human faces.

[0013] According to an embodiment of the present invention, processing digital images includes analyzing the luminous intensity of each digital image in a stack in both spatial and temporal directions to determine a combined intensity profile for the set of digital images. Based on the segmentation, pixel intensities of each digital image are analyzed in both spatial and temporal directions to obtain an intensity profile of the stacked digital image. The intensity profile information may be used to determine the frequency intensity and / or pattern of Moiré interference noise detected in each digital image and correlate them with the corresponding image exposure and / or resolution in the stack of digital images. Correlation of Moiré patterns may be performed using feature detection and matching methods such as Fourier / Wavelet transform, SURF, SIFT features, etc. Correlation may also be performed using a trained deep learning network such as a convolutional neural network (CNN).

[0014] According to an embodiment of the present invention, generating a composite digital image includes combining details from different captured digital images in the set that contain moiré patterns into a single image that represents a range of tonal values ​​within the captured scene and / or object. Furthermore, generating the composite digital image may include spatially correlating the extracted moiré pattern information detected from each digital image in the set. According to an embodiment of the present invention, generating the composite digital image includes extracting a frequency profile of the moiré pattern information. The detected and / or extracted moiré patterns from each digital image may be mapped onto a division grid, thereby identifying portions of the digital image that contain the moiré pattern information. Due to variations in exposure and / or resolution of the digital images in the set, the frequency intensity of the moiré patterns detected in each digital image may differ from one another. The composite digital image may be generated based on selecting and combining digital images and / or portions of digital images. For example, digital images and / or portions of digital images may be selected based on detected moiré pattern information that meets certain criteria, such as frequency intensities within a specific range. As such, the extracted moiré patterns detected and / or extracted from the selected portions may be spatially correlated to determine the resulting moiré frequency distribution profile in the composite digital image. Furthermore, other known methods may be used to generate the composite digital image. For example, the composite image may be generated by selecting the maximum frequency or combining both low and high frequencies of the detected moiré patterns in the time direction in a stack of digital images.

[0015] According to an embodiment of the present invention, the composite digital image is a high dynamic range image.

[0016] According to a second aspect of the present invention, there is provided a method for determining liveness of a biometric feature of a person, said method comprising: capturing a set of digital images of one or more biometric features of a person, the set of digital images comprising images captured by one or more cameras and at different resolutions and / or exposures; According to an embodiment of the first aspect, detecting Moiré pattern information from the digital image and generating a composite digital image for one or more biometric features accordingly; extracting moiré pattern information from the composite digital image; and determining the liveness of one or more biometric features captured in the digital image based on the extracted moiré pattern information.

[0017] According to an embodiment of the second aspect of the present invention, the step of determining liveness comprises the steps of extracting moiré frequency intensities from the moiré pattern information and comparing the extracted moiré frequency intensities with a liveness threshold.

[0018] According to an embodiment of the second aspect of the present invention, if the extracted moiré frequency intensities are within a first range from a liveness threshold, the digital image of one or more biometric features is validated, otherwise the digital image is rejected.

[0019] The method of the present invention may be used in a range of applications. For example, a method for detecting moiré patterns in digital images may be part of a biometric verification system. In a biometric verification system, it is important to be able to distinguish between spoofing attacks and legitimate access requests from registered users. In a spoofing attack, such as facial spoofing, an unauthorized user may attempt to gain unauthorized access by using photographs, videos, or other materials of an authorized user's face. Therefore, a biometric verification system must be able to accurately assess the liveness of captured biometric features. Moiré patterns may be used to determine the liveness of captured biometric features by evaluating their resulting frequency and / or pattern profiles. For example, capturing an object from a photograph or video will result in a different moiré pattern than capturing the same object in the real world. However, because the moiré pattern depends on the intensity profile of the captured digital image, a biometric verification system may erroneously detect moiré patterns from a registered user's photograph and / or video, thereby leading to a successful spoofing attack. Similarly, a biometric identification system may reject a legitimate biometric verification request from a registered user due to incorrect detection of moiré patterns from a live-captured digital image. For example, in images with lower intensity, i.e., darker images, moiré patterns may be less visible or their frequency may be lower, which may result in the moiré patterns not being accurately detected by the biometric verification system and leading to a successful spoofing attack. The present invention overcomes this problem by extracting profiles of moiré patterns detected from multiple images of the same subject captured at different resolutions and / or exposures. Thus, the present invention makes it possible to accurately extract the profiles and / or frequencies of moiré patterns from digital images, thereby improving the accuracy of biometric verification systems.

[0020] According to an embodiment of the second aspect of the present invention, capturing digital images includes varying the resolution and / or exposure of at least one camera within a predetermined range between successive captures of digital images of one or more biometric features. According to an embodiment of the second aspect of the present invention, the resolution and / or exposure of the camera is varied by a predetermined value for each digital image in the set. For example, the exposure and / or image resolution may be adjusted incrementally in a stepwise process, where each adjustment step is a predetermined value. For example, the exposure of each digital image may be varied by adjusting the shutter speed and / or aperture of the camera. Similarly, the image resolution may be adjusted by changing the resolution of the camera sensor.

[0021] According to an embodiment of the second aspect of the present invention, the one or more biometric features are facial features.

[0022] According to a third aspect of the present invention, there is provided a liveness detection system for biometric authentication, said system comprising: at least one camera configured to capture a digital image of a target biometric feature of a person presented for biometric authentication; According to an embodiment of the second aspect, the method further comprises:

[0023] The following drawings are provided by way of example to further explain and describe various aspects of the present invention. [Brief explanation of the drawings]

[0024] [Figure 1] 1 illustrates an example of capturing a live image of a user. [Figure 2] 1 illustrates an example of capturing a digital image of a screen or photograph showing an image of a user. [Figure 3] 1 illustrates an example of a biometric verification system according to an embodiment of the present invention. [Figure 4] 1 illustrates an example processing pipeline for detecting liveness of a biometric feature according to an embodiment of the present invention. [Figure 5] 1 illustrates an example of varying the resolution of a camera sensor according to an embodiment of the present invention. [Figure 6] 1 illustrates an example set of multivariate digital images captured at different resolutions and / or exposures according to an embodiment of the present invention. [Figure 7] 1 illustrates an example of how a captured multivariate digital image may be segmented according to an embodiment of the present invention. [Figure 8] 1 illustrates an example of how different moiré patterns can be detected in different digital images according to an embodiment of the present invention. [Figure 9] 1 illustrates an example of a composite digital image according to an embodiment of the present invention. [Figure 10] 1 illustrates an example composite image of Moiré patterns detected in a set of digital images according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0025] The present invention will be illustrated using exemplary embodiments shown in FIGS. 1-10 and described in more detail below. It should be noted that any references made to dimensions are merely exemplary and are not intended to limit the invention in any way. While the present invention has been shown and described with reference to certain exemplary embodiments thereof, it will be understood by those skilled in the art that various changes in form and detail can be made thereto without departing from the scope of the invention as encompassed by the appended claims. Furthermore, while the present invention has been described with reference to particular systems and methods for detecting moiré patterns in digital images and corresponding methods and systems for assessing the liveness of biometric features in biometric verification systems, it should be understood by those skilled in the art that the systems and methods can be used for detecting moiré patterns in other applications, such as in optical character recognition (OCR) applications, without departing from the scope of the invention as encompassed by the appended claims.

[0026] Detecting moiré patterns in digital images may be used in a range of applications. For example, detecting moiré patterns may facilitate detection of biometric feature liveness in biometric verification systems. Similarly, detecting moiré patterns may be used in OCR applications to improve detection of characters from scanned documents or photographs. With respect to liveness detection in particular, detecting moiré patterns in captured digital images has been shown to improve liveness detection accuracy. It is known that the moiré pattern detected in a captured digital image of a live subject, such as a person sitting in front of a camera, will be different from the moiré pattern detected when the captured image is a document scan or photograph of the same subject. As illustrated in FIG. 1 , capturing a digital image of a person 200 using a capture device 110, such as a mobile phone camera, may result in a detected moiré pattern having a first frequency intensity and / or profile. For example, when the image is displayed on the phone's display 111, the detected moiré pattern may be less visible and therefore have a lower intensity or a different frequency profile. However, capturing a digital image of a photograph 300 of the same subject may result in a moiré pattern having a second frequency intensity and / or profile. For example, as illustrated in Figure 2, the moiré pattern detected and / or extracted from the captured digital image of a photograph 400 displayed on the display screen 111 of a mobile phone 110 may have higher frequency intensity and / or a different profile compared to the moiré pattern detected in Figure 1. This is because the photograph 400 of a person already contains a moiré pattern 310, which may be captured and amplified by the image sensor of the capture device 110. Therefore, by detecting the moiré pattern in the captured digital image, it is possible to detect the liveness of the subject, which is particularly useful in preventing biometric spoofing attacks.

[0027] FIG. 3 illustrates an example of a biometric verification system 100 configured to verify a user's identity according to an embodiment of the present invention. The biometric verification system 100 may be configured to capture biometric features, such as facial features, via a capture module 110, e.g., a camera. The captured biometric features may be processed by a liveness detection module 120 to determine liveness. Once liveness is determined, the biometric features may be detected in a captured digital image by a biometric detection module 130 and then matched against stored biometric features by a biometric matching module 140. If a biometric match is identified, the user may be verified and may optionally be authorized to perform actions, such as accessing applications, by an authorization module 160. A communications module 150 may be provided for communicating with the user and other connected systems. The biometric verification may be aborted at any stage, for example, if liveness detection fails or no matching biometric features are found.

[0028] As before, liveness detection may be performed by detecting moiré patterns in a captured image. However, a disadvantage of known techniques is that they rely on extracting moiré patterns from a single image. It is well known that the accuracy of moiré pattern detection can be affected by pixel intensity in the captured image. For example, it has been found that the lower the pixel intensity, and therefore the darker the captured image, the lower the frequency intensity of the moiré pattern. Therefore, the accuracy of liveness detection can be significantly affected by pixel intensity in the captured image. Variations in liveness detection accuracy can lead to false positives, e.g., successful spoofing attacks, or false negatives, e.g., incorrectly rejecting a user's live biometric features. In contrast, the liveness detection system of the present invention aims to improve liveness detection accuracy by evaluating moiré patterns detected from a set of multivariate digital images taken with different image resolutions and / or exposures. In other words, the present invention varies the image resolution and / or exposure of the captured digital images to adjust the pixel intensity of the captured biometric features and / or the generated moiré patterns. By adjusting the light intensity and / or resolution of the captured image, it is possible to adjust the frequency intensity and / or profile of the moiré pattern being detected and / or extracted, thus enabling detection of moiré patterns from captured images having a range of pixel intensities.

[0029] FIG. 4 illustrates an example of a liveness detection processing pipeline according to an embodiment of the present invention. The liveness detection processing pipeline is based on detecting Moiré patterns from a set of multivariate digital images of a captured subject. In the context of the present invention, multivariate digital images refer to digital images captured by one or more image capture devices 110 at different image resolutions and / or exposures. The liveness detection processing pipeline may be part of a liveness detection module 120 of a biometric verification system 100. As illustrated in FIG. 4, an image capture device 110, such as a mobile phone camera, capable of capturing digital images may be provided. Once an authentication request is received in the biometric verification system 100, the liveness detection module 120 may operate the capture device 110 to capture a set 122 of multivariate digital images of a subject and / or scene, such as a person's facial features. For each digital image captured, the resolution and / or exposure settings of the capture device 110 may be modified by a predetermined value using a resolution / exposure modifier module 121. For example, for each digital image, the camera's aperture and / or shutter speed, which control image exposure, may be varied by a predetermined value. Similarly, the resolution of each digital image in the set may be altered, e.g., increased or decreased by a predetermined value. Image resolution and / or exposure may be varied for each digital image 122 by adjusting firmware and / or hardware settings of capture module 110. For example, resolution / exposure modifier 121 may be programmed to control capture module 110 so that each digital image is captured at a different image exposure and / or resolution, e.g., by varying shutter speed, aperture, and / or image resolution. Furthermore, capture module 110 may include multiple cameras, each configured to capture digital images at a different image resolution and / or exposure. As such, resolution / exposure modifier 121 may switch between different cameras of the capture module to capture digital images 122.Additionally, the resolution of the capture module 110 may be adjusted by providing additional hardware components in front of the camera sensor to adjust the camera sensor resolution, similar to coded photography. For example, the additional hardware component may be a mask configured to be placed in front of the camera sensor so that the camera sensor resolution is reduced. An external mask may be designed to reduce the resolution of the camera sensor when placed in front of it. An example of coded photography using hardware components is described in Raskar, Ramesh. (2009), "Computational Photography: Epsilon to Coded Photograph." As shown in FIG. 5, reducing the camera sensor 112 resolution can result in different moiré patterns 310 in the digital image, which may be used to assess the liveness of the captured subject and / or scene. The number of digital images 122 to be captured depends entirely on the application and details required. At least two images may be captured with different exposures and / or resolutions.

[0030] Returning to FIG. 4 , once the desired number of digital images in the set have been obtained, an image registration module 123 may be provided to align the captured objects and / or scenes within the set of digital images 122. Image registration may be performed as a post-processing step using any known, available algorithm. Equivalently, image registration may be performed during digital image capture, for example, using a sensor to detect and remove motion from the captured images. Once aligned, each image is segmented into portions using a segmentation module 124. The segmentation may be grid segmentation or any other known segmentation technique, such as object and / or shape detection. Based on the segmentation results, the intensity profile of each digital image is analyzed using an intensity analysis module 125. The intensity analysis module 125 is configured to analyze the light intensity of the digital images in both spatial and temporal directions using any known method, thereby generating an intensity profile for the set of digital images. Based on the intensity profile, each digital image may be correlated with a corresponding image resolution and / or exposure. Moiré patterns may be detected in each digital image by determining portions for each digital image that contain moiré pattern information. For example, a portion may be considered to contain moiré pattern information if the corresponding frequency intensity is within a predetermined range. Moiré patterns may be detected using known methods, such as wavelet techniques known in the art. Once moiré patterns are identified in each digital image 122, a moiré pattern profiling module 126 may be used to generate a profile of the moiré patterns in the set of digital images. Digital images containing moiré patterns may be combined into a single digital image, for example, using high dynamic range (HDR) techniques or similar algorithms. In this manner, associated tone values ​​and pixel intensities of related digital images may be fused together into a single image, resulting in a fusion of the moiré patterns identified at each corresponding image resolution and / or exposure.A fused image of the moiré patterns in the composite digital image may be generated using fused moiré pattern generation module 127. The fused moiré pattern image may be sent to a liveness detection algorithm, which may compare the frequency intensity and / or profile of the fused moiré pattern to a predetermined liveness threshold and / or value. The results may be sent to the remaining modules of biometric verification system 100. For example, if liveness detection returns a negative result, the captured biometric feature may be considered to have been generated from a fake image, thereby rejecting the request. Otherwise, the process may proceed to biometric detection and matching, as previously described with reference to FIG. 3.

[0031] In different applications, such as OCR applications, the processing pipeline may be adapted accordingly, such that it only applies to the detection of moiré patterns in a composite digital image, such as that produced by the Moiré Pattern Fusion Image module 128. Similarly, a more simplified process pipeline may be employed, - capturing a set of digital images of a target subject and / or scene, each taken using a different image resolution and / or exposure; - processing each digital image in the set to determine portions of the imaged object and / or scene that contain Moiré pattern information at the corresponding image resolution and / or exposure; - selecting a digital image from the set comprising portions of an imaged object and / or scene that contain moiré pattern information; - generating a composite digital image of the captured object and / or scene from the selected digital images, the composite digital image including the moiré pattern information detected in each of the selected digital images.

[0032] FIGS. 6-10 illustrate examples of how a fusion image of moiré patterns in a set of digital images can be generated. For example, as illustrated in FIG. 6, a set of three multivariate images 122 of a person's face can be captured using a camera. Each image can be captured at a different resolution and / or exposure, resulting in variations in light intensity within the set of digital images 122. For example, each digital image 122 can be captured at a different exposure time, e.g., 1 / 8, 1 / 2, or 2, while the remaining parameters, e.g., ISO and aperture, can remain the same. It should be noted that other parameters can also be adjusted as needed. Once captured, the digital images 122 can be divided into a grid of predetermined dimensions using an image division module 124, as illustrated in FIG. 7. Based on the division, a light intensity profile for each digital image 122 can be determined. Moiré patterns 310-A, 310-B, and 310-C can be detected in each digital image 122 using a moiré pattern profiling module 126, as illustrated in FIG. 8. The related digital images, in this case all three, containing moiré patterns of a certain frequency and / or profile are combined using a moiré pattern fusion module 127, as illustrated in Figure 9, to generate a composite digital image 127a containing moiré patterns from all related images 122. The composite digital image 127a is an HDR image and may therefore contain details from all related digital images 122. From the composite digital image 127a, a fused moiré pattern image 128 may be extracted, as illustrated in Figure 10.

[0033] In general, the routines executed to implement embodiments of the present invention, whether implemented as part of an operating system, or as a specific application, component, program, object, module, or sequence of instructions, or even a subset thereof, may be referred to herein as "computer program code" or simply "program code." Program code typically resides at different times in various memory and storage devices within a computer, and is comprised of computer-readable instructions that, when loaded and executed by one or more processors within the computer, cause the computer to perform the operations necessary to carry out the operations and / or elements embodying various aspects of embodiments of the present invention. Computer-readable program instructions for carrying out operations of embodiments of the present invention may be, for example, assembly language, or source code or object code written in any combination of one or more programming languages.

[0034] The program code embodied in any of the applications / modules described herein can be distributed individually or collectively as a program product in a variety of different forms. In particular, the program code may be distributed using a computer-readable storage medium having computer-readable program instructions thereon to cause a processor to implement aspects of embodiments of the present invention. Computer-readable storage media are non-transitory in nature and may include volatile and non-volatile, removable and non-removable tangible media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules, or other data. Computer-readable storage media may also include RAM, ROM, erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other robust state memory technology, portable compact disc read-only memory (CD-ROM) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device, or any other medium usable to store desired information and readable by a computer. The computer-readable storage medium should not be interpreted as a transitory signal itself (e.g., radio waves or other propagating electromagnetic waves, electromagnetic waves propagating through a transmission medium such as a waveguide, or an electrical signal transmitted through an electrical wire). The computer-readable program instructions may be downloaded from the computer-readable storage medium or an external computer or external storage device via a network to a computer, another type of programmable data processing apparatus, or another device. The computer-readable program instructions stored on the computer-readable medium may be used to direct a computer, other type of programmable data processing apparatus, or other device to function in a particular manner, and the instructions stored on the computer-readable medium may generate an article of manufacture including instructions that implement the functions / acts specified in the flowcharts, sequence diagrams, and / or block diagrams.Computer program instructions may be provided to one or more processors of a general purpose computer, special purpose computer or other programmable data processing apparatus to cause a machine, the instructions being executed by the one or more processors, to perform a sequence of calculations to implement the functions and / or acts specified in the flowcharts, sequence diagrams and / or block diagrams.

[0035] In some alternative embodiments, the functions and / or acts specified in the flowcharts, sequence diagrams, and / or block diagrams may be rearranged, processed sequentially, and / or processed simultaneously without departing from the scope of the invention. Moreover, any of the flowcharts, sequence diagrams, and / or block diagrams may include more or fewer blocks than illustrated in accordance with an embodiment of the invention.

[0036] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the embodiments of the present invention. As used herein, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context indicates otherwise. It will be further understood that the terms "comprises" and / or "comprising," as used herein, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. Furthermore, to the extent that the terms "includes," "having," "has," "with," "comprised of," or variations thereof are used in the detailed description or claims, such terms are intended to be as inclusive as the term "comprising."

[0037] While the description of various embodiments has been fully illustrative of the invention, and while these embodiments have been described in considerable detail, it is not the intention of applicants to restrict or in any way limit the scope of the appended claims to such details. Additional advantages and modifications will be readily apparent to those skilled in the art. The invention in its broader aspects is therefore not limited to the specific details, representative apparatus and methods, and illustrative examples shown and described. Accordingly, departures may be made from such details without departing from the spirit or scope of applicants' general inventive concept. [Explanation of symbols]

[0038] 100 Biometric Systems 110 Capture Module 111 Display 112 Camera Sensor 120 Liveness Detection Module 121 Resolution / Exposure Changer Module A set of 122 multivariate digital images 123 Image Registration Module 124 division module 125 Strength Analysis Module 126 Moire Pattern Profiling Module 127 Fusion Moire Pattern Generation Module 127a Composite Digital Image 128 Moire Pattern Fusion Image Module 130 Biometric Detection Module 140 Biometric Matching Module 150 Communication Module 160 Authorization Module 200 people 300 photos 310, 310-A, 310-B, 310-C Moire pattern 400 photos

Claims

1. 1. A method for detecting moiré pattern information in a digital image, comprising: receiving a set of digital images of an object and / or scene, said set of digital images including images captured by one or more cameras and at different resolutions and / or exposures; processing each digital image in the set to determine a portion of the imaged object and / or scene that contains Moiré pattern information at a corresponding image resolution and / or exposure; selecting a digital image from said set comprising a portion of said imaged object and / or scene containing Moiré pattern information; generating a composite digital image of the captured object and / or scene from the selected digital images, the composite digital image including the Moiré pattern information detected in each of the selected digital images; A method comprising:

2. The method of claim 1 , wherein processing the digital images includes compensating for motion of the subject and / or scene captured in each successive digital image in the set.

3. 3. The method of claim 1, wherein the step of processing the digital images comprises dividing each digital image into a grid of predetermined dimensions.

4. The method of claim 3 , wherein processing the digital images includes detecting grid portions in each digital image that contain Moire pattern information.

5. 5. The method of claim 4, wherein processing the digital images comprises analyzing the light intensity of each digital image in the stack in both spatial and temporal directions to determine a combined intensity profile for the set of digital images.

6. 6. The method of claim 1, wherein generating a composite digital image comprises combining different captured digital images in the set that include a moiré pattern into a single image that represents a range of tonal values ​​within the captured subject and / or scene.

7. The method of claim 1 , wherein generating a composite digital image comprises extracting a profile of the Moiré pattern information.

8. 8. The method of claim 1, wherein the composite digital image is a high dynamic range image.

9. 1. A method for determining liveness of a biometric feature of a person, comprising: capturing a set of digital images of one or more biometric features of a person, said set of digital images comprising images captured by one or more cameras and at different resolutions and / or exposures; Detecting Moiré pattern information from the digital image according to the method of any one of claims 1 to 8 and generating a composite digital image for the one or more biometric features accordingly; extracting moiré pattern information from the composite digital image; determining liveness of the one or more biometric features captured in the digital image based on the extracted Moiré pattern information; A method comprising:

10. The method of claim 9 , wherein determining liveness comprises extracting moiré frequency intensities from the moiré pattern information and comparing the extracted moiré frequency intensities to a liveness threshold.

11. 11. The method of claim 10, further comprising validating the digital image of the one or more biometric features if the extracted Moiré frequency intensities are within a first range from the liveness threshold, and rejecting the digital image otherwise.

12. 12. The method of claim 9, wherein capturing the digital images comprises varying the resolution and / or exposure of at least one camera within a predetermined range between successive captures of digital images of the one or more biometric features.

13. The method of claim 12 , wherein the camera resolution and / or exposure is varied by a predetermined value for each digital image in the set.

14. 14. The method of any one of claims 9 to 13, wherein the one or more biometric features are facial features.

15. A liveness detection system for biometric authentication, comprising: at least one camera configured to capture a digital image of a target biometric feature of a person presented for biometric authentication; a processor configured to determine the liveness of a captured biometric feature according to the method of any one of claims 9 to 14; A living body detection system comprising:

Images