Method and system for modifying a user biometric from an image

Abstract

The present disclosure relates to a method [300] and a system [100] for modifying a user biometric form an image. The method [300] includes identifying one or more exposed biometric areas associated with the user in the image. The method [300] includes determining one or usable biometric regions within each of the one or more exposed biometric areas. The method [300] further includes modifying the one or more usable biometric regions while keeping aesthetics of the image.

Description

METHOD AND SYSTEM FOR MODIFYING A USER BIOMETRIC FROM AN IMAGE

[0001] Embodiments of the present disclosure generally relate to biometric data protection and privacy technology. More particularly, embodiments of the present disclosure relate to a method and system for modifying a user biometric from an image.

[0002] The following description of the related art is intended to provide background information pertaining to the field of the disclosure. This section may include certain aspects of the art that may be related to various features of the present disclosure. However, it should be appreciated that this section is used only to enhance the understanding of the reader with respect to the present disclosure, and not as admissions of the prior art.

[0003] In recent years, integration of biometric data, particularly fingerprints, as a method of identification and authentication has become increasingly prevalent across various digital platforms. Biometric authentication provides a convenient and secure means to access devices, and secure sensitive information compared to traditional password-based approaches. However, the widespread use of high-resolution cameras in smartphones and other smart devices (such as smartwatches) has introduced significant security risks. These devices can capture intricate details, including biometric features such as fingerprints, with unprecedented clarity and accuracy. This capability presents a serious concern as the photographs / images taken in everyday scenarios, whether intentionally or incidentally, can inadvertently expose biometric data without user's consent or awareness. For example, in a world of social media, people often click images / videos with exposed biometric details (a selfie with a hand victory or peace sign) and upload on public platforms. The biometric details such as fingerprint information can be 100% restored if the image was clicked / taken within 1.5 meters (5 feet) of the camera. The exploitation of biometric details extracted from digital images poses substantial risks to individuals and organizations. Such data can be misused for identity theft, unauthorized access to personal accounts and devices, and various forms of financial fraud. Unlike passwords or PINs, which can be changed if compromised, biometric identifiers such as fingerprints are immutable, making the consequences of their exposure long-lasting and difficult to mitigate.

[0004] There are various types of biometric data such as facial biometric data, iris biometric data and fingerprint biometric data. However, fingerprint biometric is the most used biometric as facial and iris biometric data require liveliness information which may be difficult to use everywhere. The fingerprint data can be used statically (such as bank cheques, physical document verification etc.), as well as dynamically (such as liveliness enabled fingerprint optical sensors etc.).

[0005] Existing methods to mitigate these risks primarily involve applying generic image manipulation techniques such as blur filters, pixelation, or edge detection to obfuscate sensitive areas within the images. While these techniques provide a basic level of protection, they often compromise the visual quality of the image, and fail to sufficiently mask / conceal detailed biometric information. Also, these techniques may introduce artifacts that render the image less usable or realistic. Moreover, certain biometric features, such as fingerprints, lead to unique challenges due to their distinctive patterns and the potential for extraction from various surfaces (e.g., skin, glass, metal). Techniques that effectively protect such sensitive data while maintaining the integrity and usability of digital images remain a critical area of concern. Also, fingerprints can not only be extracted from normal images, but also from fingerprint art, colored hands, finger tattoos, and the like.

[0006] In view of the above-mentioned challenges, there exists a need for advanced methods and systems designed specifically to safeguard biometric data within digital images effectively while maintaining quality of the digital images. More specifically, there is a need in the art to provide an enhanced solution for modifying user biometric from an image.

[0007] This section is provided to introduce certain aspects of the present disclosure in a simplified form that are further described below in the detailed description. This summary is not intended to identify the key features or the scope of the claimed subject matter.

[0008] An aspect of the present disclosure may relate to a method to modify a user biometric from an image, the method comprising identifying one or more exposed biometric areas associated with the user in the image. The method further comprises determining, one or more usable biometric regions within each of the one or more exposed biometric areas. The method further comprises modifying, the one or more usable biometric regions while keeping aesthetics of the image.

[0009] In an exemplary aspect of the present disclosure, the image has at least one of a colour impression or an exposed fingerprint of the one or more exposed biometric areas associated with the user.

[0010] In an exemplary aspect of the present disclosure, the one or more exposed biometric areas associated with the user in the image are identified based on segmenting the image to extract at least a hand, a fingerprint and a fingertip edge.

[0011] In an exemplary aspect of the present disclosure, determining, the one or more usable biometric regions within the one or more exposed biometric areas comprises enhancing, the segmented image to extract feature points, wherein each of the feature point comprises a plurality of minutiae points. The method further comprises clustering, the plurality of minutiae points to obtain the one or more usable biometric regions, wherein each of the one or more usable biometric regions comprises a predefined number of minutiae points. The method further comprises calculating a usability score of each of the one or more usable biometric regions, based on a collective characteristic of each of the minutiae point from the plurality of minutiae points within each of the usable biometric region.

[0012] In an exemplary aspect of the present disclosure, the extracted feature points further comprise one or more user specific feature points comprising at least a mole and a scar.

[0013] In an exemplary aspect of the present disclosure, modifying, the one or more usable biometric regions comprises modifying, each of the one or more usable biometric regions using a Feature Guided Fingerprint Generative (FGFG) module, based on the usability score of each of the one or more usable biometric regions.

[0014] In an exemplary aspect of the present disclosure, each of the minutiae points within each of the usable biometric region from the one or more usable biometric regions defines at least one of a ridge, a valley, a bifurcation, a dot, a ridge ending and an island.

[0015] In an exemplary aspect of the present disclosure, the collective characteristic of each of the minutiae points comprises an assigned weight to at least a type of each minutiae point, a distance of each minutiae point from a core of a corresponding usable biometric region and a local ridge density around each minutiae point.

[0016] In an exemplary aspect of the present disclosure, the core is a central reference point in the corresponding usable biometric region from the one or more usable biometric regions.

[0017] In an exemplary aspect of the present disclosure, the local ridge density around each minutiae point is a number of ridges present in the corresponding usable biometric region from the one or more usable biometric regions, around the core.

[0018] In an exemplary aspect of the present disclosure, keeping the aesthetics of the image comprises maintaining color, texture, art, paint and image applied on the one or more usable biometric regions.

[0019] Another aspect of the present disclosure may relate to a system to modify a user biometric from an image. The system includes a processing unit and a memory unit connected to the processing unit. The processing unit is configured to identify one or more exposed biometric areas associated with the user in the image. The processing unit is configured to determine one or more usable biometric regions within each of the one or more exposed biometric areas. The processing unit is configured to modify the one or more usable biometric regions while keeping aesthetics.

[0020] The accompanying drawings, which are incorporated herein, and constitute a part of this disclosure, illustrate exemplary embodiments of the disclosed methods and systems in which like reference numerals refer to the same parts throughout the different drawings. Components in the drawings are not necessarily to scale, emphasis instead being placed upon clearly illustrating the principles of the present disclosure. Also, the embodiments shown in the figures are not to be construed as limiting the disclosure, but the possible variants of the method and system according to the disclosure are illustrated herein to highlight the advantages of the disclosure. It will be appreciated by those skilled in the art that disclosure of such drawings includes disclosure of electrical components or circuitry commonly used to implement such components.

[0021] Figure 1 illustrates an exemplary block diagram of a system for modifying a user biometric from an image, in accordance with exemplary embodiments of the present disclosure.

[0022] Figure 2A illustrates an exemplary functional block diagram of the system for modifying the user biometric from the image, in accordance with exemplary embodiments of the present disclosure.

[0023] Figure 2B illustrates an exemplary use case of the system for modifying the user biometric from the image, in accordance with exemplary embodiments of the present disclosure.

[0024] Figure 3 illustrates a flow diagram to perform a method for performing biometric data replacement for modifying the user biometric form the image, in accordance with exemplary embodiments of the present disclosure.

[0025] Figure 4 illustrates an exemplary scenario for hand detection in an input image through image segmentation, in accordance with exemplary embodiments of the present disclosure.

[0026] Figure 5 illustrates an exemplary scenario for fingertip extraction from the input image, in accordance with an exemplary embodiment of the present disclosure.

[0027] Figure 6 illustrates an exemplary use case for generating a modified image as an output, in accordance with exemplary embodiments of the present disclosure.

[0028] Figure 7A illustrates an exemplary use case for generating a skeleton image to prevent exposure of fingerprint of a user from fingerprint binary image, in accordance with an embodiment of the present disclosure.

[0029] Figure 7B illustrates an exemplary use case method for generating fake fingerprint with false minutiae features from the skeleton image, in accordance with an embodiment of the present disclosure.

[0030] Figure 8A and Figure 8B illustrate an exemplary flow diagram for generating fake fingerprint, in accordance with an embodiment of the present disclosure.

[0031] Figure 9 illustrates a flow diagram depicting a method to modify a user biometric from an image for prevention of exposed user biometric in the image from getting misused, in accordance with exemplary embodiments of the present disclosure.

[0032] The foregoing shall be more apparent from the following more detailed description of the disclosure.

[0033] In the following description, for the purposes of explanation, various specific details are set forth in order to provide a thorough understanding of embodiments of the present disclosure. It will be apparent, however, that embodiments of the present disclosure may be practiced without these specific details. Several features described hereafter may each be used independently of one another or with any combination of other features. An individual feature may not address any of the problems discussed above or might address only some of the problems discussed above.

[0034] The ensuing description provides exemplary embodiments only, and is not intended to limit the scope, applicability, or configuration of the disclosure. Rather, the ensuing description of the exemplary embodiments will provide those skilled in the art with an enabling description for implementing an exemplary embodiment. It should be understood that various changes may be made in the function and arrangement of elements without departing from the spirit and scope of the disclosure as set forth.

[0035] Specific details are given in the following description to provide a thorough understanding of the embodiments. However, it will be understood by one of ordinary skill in the art that the embodiments may be practiced without these specific details. For example, circuits, systems, processes, and other components may be shown as components in block diagram form in order not to obscure the embodiments in unnecessary detail.

[0036] Also, it is noted that individual embodiments may be described as a process which is depicted as a flowchart, a flow diagram, a data flow diagram, a structure diagram, or a block diagram. Although a flowchart may describe the operations as a sequential process, many of the operations may be performed in parallel or concurrently. In addition, the order of the operations may be re-arranged. A process is terminated when its operations are completed but could have additional steps not included in a figure.

[0037] The word "exemplary" and / or "demonstrative" is used herein to mean serving as an example, instance, or illustration. For the avoidance of doubt, the subject matter disclosed herein is not limited by such examples. In addition, any aspect or design described herein as "exemplary" and / or "demonstrative" is not necessarily to be construed as preferred or advantageous over other aspects or designs, nor is it meant to preclude equivalent exemplary structures and techniques known to those of ordinary skill in the art. Furthermore, to the extent that the terms "includes," "has," "contains," and other similar words are used in either the detailed description or the claims, such terms are intended to be inclusive in a manner similar to the term "comprising" as an open transition word without precluding any additional or other elements.

[0038] As used herein, "a user equipment", "a user device", "a smart-user-device", "a smart-device", "an electronic device", "a mobile device", "a handheld device", "a mobile communication device", or the like may be any electrical, electronic and / or computing device or equipment, capable of implementing one or more features of the present disclosure. The user equipment / device may include, but is not limited to, a mobile phone, smart phone, laptop, a general-purpose computer, desktop, personal digital assistant, tablet computer, wearable device or any other computing device which is capable of implementing the features of the present disclosure.

[0039] As used herein, "storage unit" or "memory" refers to a machine or computer-readable medium including any mechanism for storing information in a form readable by a computer or similar machine. For example, a computer-readable medium includes read-only memory ("ROM"), random access memory ("RAM"), magnetic disk storage media, optical storage media, flash memory devices or other types of machine-accessible storage media. The storage unit stores at least the data that may be required by one or more units of the system to perform their respective functions.

[0040] All modules, units, components used herein, unless explicitly excluded herein, may be software modules or hardware processors, the processors being a general-purpose processor, a special purpose processor, a conventional processor, a digital signal processor (DSP), a plurality of microprocessors, one or more microprocessors in association with a DSP core, a controller, a microcontroller, Application Specific Integrated Circuits (ASIC), Field Programmable Gate Array circuits (FPGA), any other type of integrated circuits, etc.

[0041] One or more of the plurality of modules may be implemented through an AI model. A function associated with AI may be performed through the non-volatile memory, the volatile memory, and the processor. The processor may include one or a plurality of processors. For implementing the one or the plurality of modules through an AI model, the one or the plurality of processors may be a general purpose processor(s), such as a central processing unit (CPU), an application processor (AP), or the like, a graphics-only processing unit such as a graphics processing unit (GPU), a visual processing unit (VPU), and / or an AI-dedicated processor such as an image processor. The one or the plurality of processors control the processing of the input data in accordance with a predefined operating rule or artificial intelligence (AI) model stored in the non-volatile memory and the volatile memory. The predefined operating rule or artificial intelligence model is provided through training or learning. Here, being provided through learning means that, by applying a learning algorithm(s) to a plurality of learning data, a predefined operating rule or AI model of a desired characteristic is made. The learning may be performed in a device itself in which AI according to an embodiment is performed, and / or may be implemented through a separate server / system. The AI model may consist of a plurality of neural network layers, such as long short-term memory (LSTM) layers. Each layer has a plurality of weight values and performs a layer operation through calculation of a previous layer and an operation of a plurality of weights. Examples of neural networks include, but are not limited to, convolutional neural network (CNN), deep neural network (DNN), recurrent neural network (RNN), restricted Boltzmann Machine (RBM), deep belief network (DBN), bidirectional recurrent deep neural network (BRDNN), generative adversarial networks (GAN), and deep Q-networks.

[0042] As discussed in the background section, Biometric authentication provides a convenient and secure means to access devices, and secure sensitive information compared to traditional password-based approaches. However, the widespread use of high-resolution cameras in smartphones and other smart devices (such as smartwatches) has introduced significant security risks. These devices can capture intricate details, including biometric features such as fingerprints, with unprecedented clarity and accuracy. This capability presents a serious concern as the photographs / images taken in everyday scenarios, whether intentionally or incidentally, can inadvertently expose biometric data without the user's consent or awareness. For example, in a world of social media, people often click images / videos with exposed biometric details (a selfie with a hand victory or peace sign) and upload on public platforms. The biometric details such as fingerprint information can be 100% restored if the image was clicked / taken within 1.5 meters (5 feet) of the camera. The exploitation of biometric details extracted from digital images poses substantial risks to individuals and organizations. Such data can be misused for identity theft, unauthorized access to personal accounts and devices, and various forms of financial fraud. Unlike passwords or PINs, which can be changed if compromised, biometric identifiers such as fingerprints are immutable, making the consequences of their exposure long-lasting and difficult to mitigate.

[0043] The present disclosure tends to overcome the above-mentioned challenges by using a Feature Guided Fingerprint Generative (FGFG) mode. Hereinafter, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present disclosure.

[0044] Referring to Figure 1, an exemplary block diagram of a system

[0100] for modifying a user biometric from an image is shown, in accordance with exemplary embodiments of the present disclosure. In an embodiment, the user biometric may be a fingerprint. In addition, the image may be captured from at least a camera or a digital media. The digital media includes but may not be limited to social media, websites, web pages, digital videos, electronic documents, advertisements, digital art, virtual reality, and video games. The system

[0100] includes exemplary modules to implement one or more features of the present disclosure.

[0045] As shown in Figure 1, the exemplary modules of the system

[0100] may be a processing unit

[0102] , a memory unit

[0104] , an image processing module

[0106] , a segmentation module

[0108] , and a Feature Guided Fingerprint Generative (FGFG) module

[0110] . Each of these modules may be explained in detail with reference to one or more figures in the forthcoming description. Further, for modifying the user biometric from the image, other associated software / hardware components may also be used in conjunction with the system

[0100] . The exemplary modules in Figure. 1, in an implementation, may be implemented by the processing unit

[0102] of the system

[0100] as shown in Figure 1. In an embodiment, the processing unit

[0102] may be a general-purpose processor(s), such as a central processing unit (CPU), an application processor (AP), or the like, a graphics-only processing unit such as a graphics processing unit (GPU), a visual processing unit (VPU), and / or an AI-dedicated processor such as a neural processing unit (NPU). Further, the processing unit

[0102] is associated with the memory unit

[0104] . The memory unit

[0104] stores one or more instructions for the processing unit

[0102] and other modules of the system

[0100] to implement the one or more features of the present disclosure. In an embodiment, the memory unit

[0104] stores at least data that may be required by the exemplary modules of the system

[0100] to perform their respective functions.

[0046] Further, the image processing module

[0106] receives the image as input from the camera or the digital media. The image processing module

[0106] is associated with the segmentation module

[0108] . The image processing module

[0106] processes the image to identify one or more exposed biometric areas associated with the user in the image via the segmentation module

[0108] . The segmentation module

[0108] is configured to segment the image to extract at least a hand, a fingerprint and a fingertip edge. The one or more exposed biometric areas associated with the user in the image are identified based on segmenting of the image. In an embodiment, the image has at least one of a colour impression or an exposed fingerprint of the one or more exposed biometric areas associated with the user. In an example, the image may be a selfie of the user with a hand victory sign that exposes fingerprints of two fingers of the user. In the present example, the one or more exposed biometric areas in the image are the fingerprints of the two fingers of the user. In another example, the image may be a picture of a user showing coloured hands.

[0047] Based on identification of the one or more exposed biometric areas, the system

[0100] is configured to determine one or more usable biometric regions within each of the one or more exposed biometric areas using a biometric usability module

[0206] as shown in Figure 2A. The one or more exposed biometric areas are further modified by the FGFG module

[0110] (explained in detail in Figure 2A).

[0048] Figure 2A illustrates an exemplary functional block diagram [200A] of the system

[0100] for modifying the user biometric from an image

[0202] , in accordance with exemplary embodiments of the present disclosure. The image

[0202] is same as the image described in Fig. 1. The image processing module

[0106] receives the image

[0202] as an input for image processing. The image

[0202] may be interchangeably used with the input

[0202] . After image processing, the image processing module

[0106] may send the image

[0202] to the segmentation module

[0108] to identify the one or more exposed biometric areas associated with the user in the image

[0202] .

[0049] The segmentation module

[0108] receives a data component

[0204] for segmenting the image

[0202] to identify the one or more exposed biometric areas. The data component

[0204] is a field that manipulates the way data needs to be presented, saved, segmented or organized. The data herein refers to as the image

[0202] . Further, the segmentation module

[0108] is configured to segment the image

[0202] into one or more segments to identify the one or more exposed biometric areas. The segmentation module

[0108] segments the image

[0202] to extract at least the hand, the fingerprint, and the fingertip edge. To segment the image

[0202] , the segmentation module

[0108] is configured to partition the image

[0202] into small image segments or groups for detection or extraction of the hand, the fingerprint, or the fingerprint edge present in the image

[0202] . The segmentation module

[0108] sends the segmented image

[0202] to a biometric usability module

[0206] of the system

[0100] .

[0050] The biometric usability module

[0206] is configured to determine the one or more usable biometric regions within each of the one or more exposed biometric areas. In an example, the one or more usable biometric regions refers to specific regions with sufficient detail and distinctiveness. For instance, the one or more usable biometric regions may be the region on any finger that have a clear ridge pattern. In another example, let us assume that an index finger has a central region with a distinct ridge ending and bifurcation pattern. The central region may be determined as a usable biometric region.

[0051] The biometric usability module

[0206] includes an image enhancement module

[0208] , a feature extraction module

[0210] , a region detection module

[0212] , and a usability score module

[0214] . To determine the one or more usable biometric regions, the image enhancement module

[0208] is configured to enhance the segmented image

[0202] for the feature extraction module

[0210] to extract feature points of the segmented image

[0202] . Each of the feature points comprises a plurality of minutiae points. Each of the plurality of minutiae points within each of the usable biometric region from the one or more usable biometric regions defines at least one of a ridge, a valley, a bifurcation, a dot, a ridge ending and an island. Generally, in a fingerprint image, ridges appear as dark lines while valleys are light areas between the ridges. Additionally, minutiae points are the locations where a ridge becomes discontinuous. A ridge can either come to an end, which is referred to as termination or it can split into two ridges, which is referred to as bifurcation. In an embodiment of the present disclosure, the plurality of minutiae points may be used to determine uniqueness of a fingerprint extracted from the segmented image

[0202] . The extracted feature points further comprise one or more user specific feature points. The one or more user specific feature points comprises at least a mole and a scar. In an embodiment, the image enhancement module

[0208] enhances the segmented image

[0202] for the feature extraction module

[0210] to extract maximum feature points.

[0052] Further, the region detection module

[0212] of the biometric usability module

[0206] is configured to cluster the plurality of minutiae points to obtain the one or more usable biometric regions. Each of the one or more usable biometric regions comprises a predefined number of minutiae points. In an example, the region detection module

[0212] is used to divide a minutiae map into the one or more usable biometric regions which includes at least 10 minutiae points. Further, a graph G is generated based on the 10 minutiae points where Graph G = (V, E) (as shown in Figure 2A in the region detection module

[0212] ) using a Louvain algorithm. The Louvain algorithm is a graph-based algorithm which is used to cluster the minutiae points into the matching cluster (matching cluster is the cluster which can be used for a successful match).

[0053] In the graph G, "V" represents set of vertices representing the minutiae point, wherein each vertex corresponds to a minutia point and contains attributes such as co-ordinates, type, and orientation of the minutia point. "E" represents set of edges representing spatial connection between two minutiae points, which represents the spatial proximity and potential ridge flow. Each vertex is assigned to its own community. Iteratively vertices are moved between communities to maximize modularity of the network. Each resulting community does not exceed 10 vertices. If a community exceeds more than 10 vertices, the region detection module

[0212] is configured to split it into smaller communities. Further, any cluster present in the minutiae graph G represents the usability of the biometric data.

[0054] For each of the one or more usable biometric regions, the usability score calculation module is configured to calculate a usability score. The usability score is calculated based on a collective characteristic of each of the minutiae point from the plurality of minutiae points within each of the usable biometric region. To find the usability score of the whole region detected during region detection, the usability score of each of the plurality of minutiae points within the region may be summed up together.

[0055] In an example, for a cluster of minutiae points, the usability score may be higher if it has a well-defined central core, a high density of ridges around it, and a balanced distribution of various minutiae points. The collective characteristic of each of the minutiae points comprises an assigned weight to at least a type of each minutia point, a distance of each minutiae point from a core of a corresponding usable biometric region and a local ridge density around each minutiae point. The assigned weight is utilized to prioritize the importance of different types of minutiae points of the plurality of minutiae points. In an embodiment, each of the plurality of minutiae points is weighted based on importance and contribution of each of the plurality of minutiae points towards the overall usable biometric region. The core is a central reference point in the corresponding usable biometric region from the one or more usable biometric regions. In addition, the local ridge density around each minutiae point is a number of ridges present in the corresponding usable biometric region from the one or more usable biometric regions, around the core.

[0056] In an exemplary scenario, consider that the identified ridge endings in a fingerprint image is 20 and bifurcations are 15. The minutiae points are grouped into clusters based on their proximity and spatial arrangement. Suppose three clusters (Cluster 1, Cluster 2, and Cluster 3) are created, each representing different parts of the fingerprint. For each cluster, usability score is calculated based on minutiae point types and their characteristics such as assigned weights, distance from core, and local ridge density. Let us assume that ridge endings have assigned weight of W1, and bifurcations have assigned weight of W2. The average distance of minutiae points from a central core within each cluster is calculated. Further, number of ridges surrounding each minutia point in each cluster is counted. Let us now assume that Cluster 1 contains 10 ridge endings and 5 bifurcations, with high local ridge density and a short average distance from the core. Hence, the Custer 1 may have a high usability score. Cluster 2 contains 5 ridge endings and 10 bifurcations, with moderate local ridge density and a medium distance from the core. Hence, the Cluster 2 may have a moderate usability score. Cluster 3 contains 5 ridge endings and no bifurcations, with low local ridge density and a long average distance from the core. Hence, the Cluster 3 may have a low usability score.

[0057] Further, outputs of the image enhancement module

[0208] (enhanced segmented image), the feature extraction module

[0210] (extracted feature points), the region detection module

[0212] (one or more usable biometric regions), and the usability score calculation module

[0214] are sent to the FGFG module

[0110] . The FGFG module

[0110] is configured to modify the one or more usable biometric regions while keeping aesthetics of the image

[0202] . In an embodiment, keeping the aesthetics of the image comprises maintaining colour, texture, art, paint and image applied on the one or more usable biometric regions. Further, in an implementation, the one or more usable biometric regions are modified using the FGFG module

[0110] , based on the usability score of each of the one or more biometric regions. Further, the FGFG module

[0110] generates a modified image

[0216] as an output. In an embodiment, maintaining colour of the image

[0202] in the output or modified image

[0216] refers to preserving overall color scheme or hue of the image

[0202] , during modification. In an exemplary embodiment, FGFG module

[0110] is a type of Generative Adversarial Network (GAN) tailored for generating or modifying biometric features, specifically fingerprints. The FGFG module

[0110] is configured to handle complexities of the user biometric, ensuring that the modified biometric features appear realistic.

[0058] Figure 2B illustrates an exemplary use case of the system

[0100] for modifying the user biometric from the input image

[0202] , in accordance with exemplary embodiments of the present disclosure. The input image

[0202] is processed through the image processing module

[0106] as explained in Figure 2A. The input image

[0202] may be received from a camera or any digital media having colour impression and exposed fingerprint. Further, the processed image

[0202] is passed through the biometric usability module

[0206] to identify most potential areas (one or more usable biometric regions) that may be misused. The biometric usability module

[0206] includes the image enhancement module

[0208] , the feature extraction module

[0210] , the region detection module

[0212] , and the usability score calculation module

[0214] . It is to be noted that the description of the modules

[0208] -

[0214] of the biometric usability module

[0206] is not described here as it has already been described in Figure 2A. The most potential areas are taken as input by the FGFG module

[0110] that is configured to modify the image

[0202] with usable fingerprint information and generate the modified image as an output

[0216] . The output

[0216] is generated with modified fingerprint and the aesthetics of the output

[0216] remain same as the input image

[0202] .

[0059] Figure 3 illustrates a flow diagram to perform a method

[0300] for performing biometric data replacement for modifying the user biometric form the image

[0202] , in accordance with exemplary embodiments of the present disclosure. The method initiates at step

[0302] .

[0060] Following step

[0302] , at step

[0304] , the method

[0300] includes identifying the one or more exposed biometric areas through image segmentation. The image segmentation is performed by the segmentation module

[0108] as described in Figure 1. Further, the segmentation module

[0108] is configured to segment the image

[0202] into one or more segments to identify the one or more exposed biometric areas. The segmentation module

[0108] segments the image

[0202] to extract at least the hand, the fingerprint, and the fingertip edge. To segment the image

[0202] , the segmentation module

[0108] is configured to partition the image

[0202] into small image segments or groups for detection or extraction of the hand, the fingerprint, or the fingerprint edge present in the image

[0202] .

[0061] At step

[0306] , the method

[0300] includes utilizing the biometric usability module

[0206] for image enhancement, feature extraction, region detection and usability score calculation. For image enhancement, the image enhancement module

[0208] is configured to enhance the segmented image

[0202] for the feature extraction module

[0210] to extract feature points of the segmented image

[0202] . Each of the feature points comprises a plurality of minutiae points. Each of the plurality of minutiae points within each of the usable biometric region from the one or more usable biometric regions defines at least one of a ridge, a valley, a bifurcation, a dot, a ridge ending and an island (as explained in Figure 2A). In an embodiment of the present disclosure, the plurality of minutiae points may be used to determine uniqueness of a fingerprint extracted from the segmented image

[0202] . The extracted feature points further comprise one or more user specific feature points. The one or more user specific feature points comprises at least a mole and a scar. In an embodiment, the image enhancement module

[0208] enhances the segmented image

[0202] for the feature extraction module

[0210] to extract maximum feature points.

[0062] Further, for region detection, the region detection module

[0212] of the biometric usability module

[0206] is configured to cluster the plurality of minutiae points to obtain the one or more usable biometric regions. Each of the one or more usable biometric regions comprises a predefined number of minutiae points. In an example, the region detection module

[0212] is used to divide a minutiae map into the one or more usable biometric regions which includes at least 10 minutiae points. A graph G is generated based on the 10 minutiae points where Graph G = (V, E) (as shown in Fig. 2A in the region detection module

[0212] ). For each of the one or more usable biometric regions, the usability score calculation module

[0214] is configured calculate a usability score. The usability score is calculated based on a collective characteristic of each of the minutiae point from the plurality of minutiae points within each of the usable biometric region. In an example, for a cluster of minutiae points, the usability score may be higher if it has a well-defined central core, a high density of ridges around it, and a balanced distribution of various minutiae points. The collective characteristic of each of the minutiae points comprises an assigned weight to at least a type of each minutiae point, a distance of each minutiae point from a core of a corresponding usable biometric region and a local ridge density around each minutiae point. The assigned weight is utilized to prioritize the importance of different types of minutiae points of the plurality of minutiae points. In an embodiment, each of the plurality of minutiae points is weighted based on importance and contribution of each of the plurality of minutiae points towards the overall usable biometric region. The core is a central reference point in the corresponding usable biometric region from the one or more usable biometric regions. In addition, the local ridge density around each minutiae point is a number of ridges present in the corresponding usable biometric region from the one or more usable biometric regions, around the core.

[0063] At step

[0308] the method

[0300] includes utilizing the FGFG module

[0110] to generate a variation of biometric segment for the image

[0202] . At step

[0310] , the method

[0300] includes performing biometric data replacement on the image

[0202] to generate the modified image

[0216] based on the variation of the biometric segment.

[0064] The method

[0300] terminates at step

[0316] .

[0065] Figure 4 illustrates an exemplary scenario

[0400] for hand detection in an input image

[0402] through image segmentation, in accordance with exemplary embodiments of the present disclosure. The input image

[0402] is a coloured image. The input image

[0402] is processed and converted to a grey-scaled image

[0404] using a pixel-based skin tone detection approach. For pixel-based skin tone detection, illumination component is removed from the image

[0402] to obtain an illumination invariant colour classifier, and a grey scale map of the input image

[0402] is determined that is further used to generate a grey-scaled image

[0404] . Furthermore, skin colour segmentation is performed on the grey-scale image to generate an image

[0406] . The image

[0406] is an image with highlighted biometric exposure areas. Further, from the image

[0406] , hand detection is performed and an image

[0408] is generated. The image

[0408] is an image of a detected hand.

[0066] Figure 5 illustrates an exemplary scenario

[0500] for fingertip extraction from an input image

[0502] , in accordance with an exemplary embodiment of the present disclosure. The input image

[0502] is a coloured image and may also be known as RGB image. In general, the RGB image is referred to as a TrueColor image that defines red, green and blue colour components for each individual pixel of the image. The input image

[0502] is an image of a hand. Finger detection algorithm is used to associate "left" and "right" edges to fingers (edge-pairing) and to extract the edges by their respective tips and angles.

[0067] Further, an edge pairing algorithm is used to generate an image

[0508] . The edge pairing algorithm includes hand palm segmentation where the input image

[0502] is segmented to extract finger edges. After hand palm segmentation, a segmented palm image

[0504] is generated. Also, the edge pairing algorithm includes conversion of segmented palm image

[0504] into a contour line image

[0506] showing a plurality of line segments. The segmented palm image

[0504] may be a binary blob image. To process and examine blobs contained in the binary input image

[0502] efficiently, the blobs are converted into contour lines. A contour is a sequence of pixels located along the boundaries of the blob.

[0068] Further, line segmentation of the contour lines is performed. Generally, the most important recognition features of fingers are approximately long lines along them. In order to detect such lines and deduce the presence of fingers, it is necessary to isolate them from the contour line of a blob by isolating line segments within a contour. Line segments are detected by analyzing the variation of a tangent angle on the contour. The contour line image

[0506] is an image with line segments. Further, formation of fingers based on the previously computed line segments is provided by edge pairing. A pair of edges is consequently interpreted as the "left" and "right" edge of the fingers, respectively. The line segments used for reconstruction of paired edges are plotted in the same color. An edge can only be combined with another edge. Unpaired edges are discarded and are not recognized as a finger. Finally, an image

[0508] is generated. The image

[0508] is an image with paired edges. Also, fingertip extraction algorithm is used based on assigning associated edges through pairing.

[0069] Figure 6 illustrates an exemplary use case

[0600] for generating a modified image as an output, in accordance with exemplary embodiments of the present disclosure. At step

[0602] , a coloured image is captured with biometric exposure. The coloured image is processed to remove noise by convolving the coloured image with Gaussian kernel to obtain gradient magnitude image. At step

[0604] , the gradient magnitude image is generated. The gradient magnitude image is a grey scaled image. The coloured image is grey scaled as described in Figure 4.

[0070] At step

[0606] , an edge image is generated. At step

[0608] , image segmentation is performed on the edge image. Non-maximum suppression is applied to the gradient magnitude image to thin edges to single-pixel width using an edge thinning approach. In edge thinning approach, pixels in the gradient magnitude image that are above a high threshold are considered strong edges, and pixels between a high and low threshold are considered weak edges. Image segmentation involves Convolution Neural Network (CNN)-based image segmentation that includes training a neural network to learn features from input images and predict segmentation masks that separate foreground objects from the background. In an embodiment, the image segmentation relies on convolutional layers for feature extraction, down sampling layers for spatial dimension reduction, and output layers with appropriate activation functions for semantic segmentation tasks. At step

[0610] , a binary image of fingerprint is generated. At step

[0612] , an output of fingerprint binary image is generated. The fingerprint binary image is an image of fingerprint without background.

[0071] Figure 7A illustrates an exemplary use case [700A] for generating a skeleton image

[0706] to prevent exposure of fingerprint of a user from a fingerprint binary image

[0702] , in accordance with an embodiment of the present disclosure. The fingerprint binary image

[0702] is a similar image of fingerprint without background as mentioned in Figure 6. At step 1, the fingerprint binary image

[0702] is taken as input. At step 2, the fingerprint binary image

[0702] undergoes image enhancement and an enhanced fingerprint binary image

[0704] is generated. The enhanced fingerprint binary image

[0704] is generated using the fingerprint binary image

[0702] with facilitation of the image enhancement module

[0208] as explained in Figure 2A. Further, at step 3, the enhanced fingerprint binary image

[0704] undergoes morphological skeletonization. In general, skeletonization is a process for reducing foreground regions in a binary image to a skeletal remnant that largely preserves the extent and connectivity of the original region while throwing away most of the original foreground pixels. In an embodiment, the morphological skeletonization includes dilation and erosion of the enhanced fingerprint binary image

[0704] . The dilation adds pixels to the boundaries of objects in an image and the erosion removes pixels on object boundaries. Based on the morphological skeletonization, at step 4, the skeleton image

[0706] is generated. The skeleton image

[0706] is further used to extract minutiae features (further shown in Figure 7B).

[0072] Figure 7B illustrates an exemplary use case method [700B] for generating fake fingerprint with false minutiae features from the skeleton image

[0706] , in accordance with an embodiment of the present disclosure. It is to be noted that Figure 7B is intended to be read in continuation with Figure 7A.

[0073] At step 5, the method [700B] includes utilizing the skeleton image

[0708] via the FGFG module

[0110] . At step 6, the method [700B] includes detecting endpoints and bifurcations in the skeleton image

[0708] . Based on the detected endpoints and bifurcations, the method [700B], at step 7, includes filtering false minutiae points in the skeleton image

[0706] .

[0074] At step 8, the method [700B] includes utilizing Minutia Orientation Angle (MOA) extractor and Ridge Count Direction (RCD) extractor. The MOA extractor is used to determine orientation of a plurality of minutiae points within a fingerprint image (skeleton image

[0706] ). Generally, minutiae points are key features in fingerprints, such as ridge endings and bifurcations, which are crucial for fingerprint recognition and verification. The MOA extractor may generate an orientation map that assigns an angle to each minutia point. The orientation map helps in aligning and comparing fingerprints by providing a consistent reference for ridge direction. Further, the RCD extractor is used to analyse and record direction of ridge counts in a fingerprint image. In one implementation, the RCD extractor is configured to identify ridges in the fingerprint image by detecting and tracking continuous ridges in the fingerprint image. Further, for each region of interest in the fingerprint image, the RCD extractor counts the number of ridges in specific directions by analysing segments of the fingerprint image and counting the ridges within those segments. The RCD extractor is further configured to determine the direction in which ridges are counted, which provides insight into the ridge flow and pattern in different areas of the fingerprint.

[0075] At step 9, the method [700B] includes extracting minutiae features based on the analysis performed by the MOA extractor and the RCD extractor. An image with extracted minutiae features is generated based on the extracting minutiae features. Further, based on the image with extracted minutiae features, a fake fingerprint image is generated as output (explained in detail in Figure 8A and Figure 8B).

[0076] Figure 8A and Figure 8B illustrate an exemplary flow diagram [800A, 800B (respectively)] depicting a method for generating a fake fingerprint image, in accordance with an embodiment of the present disclosure.

[0077] Referring to Figure 8A, an input image

[0802] with minutiae points is utilized to generate the fake fingerprint image as an output. At step 1, the method includes global feature extraction of the input image

[0802] . The feature extraction involves extracting features such as loops, whorls, arch, moles and scars from the input image

[0802] . Further, at step 2, the method includes K-means clustering of the input image

[0802] to obtain regions with a plurality of minutiae points. In an example, clustering of the input image

[0802] is performed to obtain regions with 6 minutiae points. Further, at step 3, the method includes usability score calculation for each of the minutiae points (as explained above in Figure 2A and Figure 2B). At step 4, the method includes utilizing a local region extractor (similar to RCD extractor explained in Figure 7B) to extract features such as radial loops, ulnar loops, double loops, plain whorls, central pocket loop whorl, accidental whorl, plain arch, and tented arch.

[0078] In general, radial loop is a fingerprint pattern where ridges enter from one side of the finger, recurve, and exit from the same side they entered, forming a pattern that flows towards the thumb side (radial side) of the hand. Radial loops are common in the index finger. In addition, an ulnar loop is a fingerprint pattern where the ridges enter from one side of the finger, recurve, and exit from the opposite side they entered, flowing towards the little finger side (ulnar side) of the hand. A double loop is a fingerprint pattern characterized by two separate loop formations within the same fingerprint impression. A plain whorl is a fingerprint pattern where the ridges form circular or spiral patterns around a central point. A central pocket loop whorl is a variation of the whorl pattern where one or more ridges make a complete circuit around the core, but the circular pattern has an enclosed area or pocket at the centre. An accidental whorl is a fingerprint pattern that does not fit the definitions of other common patterns (loop, whorl, arch) and exhibits a combination of two or more patterns. A plain arch is a fingerprint pattern where ridges enter on one side of the finger, rise in the centre, and exit on the opposite side, forming a simple arch shape. Further, a tented arch is a fingerprint pattern similar to a plain arch but with more prominent or sharper ridges in the centre, creating a tent-like appearance.

[0079] Further, the local region extractor extracts one or more features of the fingerprint image such as ending, bifurcation, line-unit, line-fragment, eye, hook, pores, line shape, incident ridges, creases and warts.

[0080] At step 5, conditional information and latent space noise input is applied on the fingerprint image with the extracted features, based on the extracted features. The conditional information includes but may not be limited to class labels, transformation parameters, along with latent space input which comprises random noise vectors sampled from a Gaussian distribution. The noise vectors introduce stochasticity in image generation.

[0081] At step 6, the method includes utilizing a variational auto-encoder generator. The variational auto-encoder generator is configured to initiate generation of fake fingerprint for each region ensuring that no minutiae points match with the original fingerprint. For generating the fake fingerprint, the FGFG module

[0110] is trained with additional loss term that penalizes local features to great extent and global features to less extent. Further at step 7, a region integration module is utilized to combine each fingerprint region such that overall structure of fingerprint may not deform. At step 8, a discriminator is utilized that is configured to minimize loss for real fingerprint and maximize loss for fake fingerprint data. At step 9, output of fake fingerprint image is generated.

[0082] Referring to Figure 8B, an input image

[0802] with minutiae points is utilized to generate the fake fingerprint image as an output (similar to Figure 8A). In addition, a noise vector and handcrafted features are inserted to the input image. The handcrafted features potentially enhance ability of the FGFG module

[0110] to generate realistic fingerprints by providing additional context / constraints based on handcrafted features. In general, the handcrafted features are manually designed, domain-specific features that are extracted from images to capture relevant information for a particular task. It is to be noted that description of additional functionality of Figure 8B is not repeated here as it has already been described in Figure 8A.

[0083] Figure 9 illustrates a flow diagram depicting a method

[0900] to modify a user biometric from an image for prevention of exposed user biometric in the image from getting misused, in accordance with exemplary embodiments of the present disclosure.

[0084] The method

[0900] initiates at step

[0902] . Following step

[0902] , at step

[0904] , the method

[0900] includes identifying one or more exposed biometric areas associated with a user in an image. The image has at least one of a color impression or an exposed fingerprint of the one or more exposed biometric areas associated with the user. The one or more exposed biometric areas associated with the user in the image are identified based on segmenting the image to extract at least a hand, a fingerprint and a fingertip edge image.

[0085] At step

[0906] , the method

[0900] includes determining one or more usable biometric regions within each of the one or more exposed biometric areas. The determining of the one or more usable biometric regions within the one or more exposed biometric areas comprises enhancing the segmented image to extract feature points. Each of the feature point comprises a plurality of minutiae points. The extracted feature points further comprise one or more user specific feature points comprising at least a mole and a scar. Further, the determining includes clustering the plurality of minutiae points to obtain the one or more usable biometric regions. Each of the one or more usable biometric regions comprises a predefined number of minutiae points. The determining further includes calculating a usability score of each of the one or more usable biometric regions, based on a collective characteristic of each of the minutiae point from the plurality of minutiae points within each of the usable biometric region. Each of the minutiae points within each of the usable biometric region from the one or more usable biometric regions defines at least one of a ridge, a valley, a bifurcation, a dot, a ridge ending and an island

[0086] At step

[0908] , the method includes modifying the one or more usable biometric regions while keeping aesthetics. In addition, modifying the one or more usable biometric regions include modifying, each of the one or more usable biometric regions using a Feature Guided Fingerprint Generative (FGFG) module

[0110] , based on the usability score of each of the one or more usable biometric regions.

[0087] The method

[0900] terminates at step

[0910] .

[0088] Some of the objects of the present disclosure, which at least one embodiment disclosed herein satisfies are listed herein below.

[0089] It is an object of the present disclosure to provide a solution for exposed biometrics in digital images from getting misused across various online and offline platforms.

[0090] It is an object of the present disclosure to generate a modified image with unusable fingerprints information from the original image through a Feature Guided Fingerprint Generative (FGFG) module keeping the quality of the image intact to prevent misuse of the exposed biometrics from the original image.

[0091] The present disclosure further discloses a non-transitory computer readable storage medium storing instruction for modifying a user biometric from an image, the instructions include executable code which, when executed by one or more units of a system

[0100] , causes a processing unit

[0102] of the system

[0100] to identify one or more exposed biometric areas associated with the user in the image. The instructions which, when executed by the one or more units of the system

[0100] causes the processing unit

[0102] to determine one or more usable biometric regions within each of the one or more exposed biometric areas and modify the one or more usable biometric regions while keeping aesthetics.

[0092] Thus, the present invention provides a novel solution for modifying user biometric in an image for prevention of exposed biometric in digital images from getting misused across various platforms. Further, the preset disclosure discloses a technically advanced solution that prevents identity theft, unauthorized access, and lack of revocability. Also, the technically advanced solution facilitates generation of fingerprint biometric image through optimized Feature Guided Fingerprint Generative module which can even work on fingerprints with art also and keeping the quality of the image intact to prevent the exposed biometrics from misuse. Further, another technical advantage of the solution lies in its ability to provide the final image with original fingerprint replaced with a fake fingerprint, with the look / colour of the image kept intact. Thus, aesthetics is same as the input image.

[0093] While considerable emphasis has been placed herein on the preferred embodiments, it will be appreciated that many embodiments can be made and that many changes can be made in the preferred embodiments without departing from the principles of the invention. These and other changes in the preferred embodiments of the invention will be apparent to those skilled in the art from the disclosure herein, whereby it is to be distinctly understood that the foregoing descriptive matter to be implemented merely as illustrative of the invention and not as limitation.

Claims

1.A method [300] for modifying a user biometric from an image, the method comprising:identifying one or more exposed biometric areas associated with the user in the image;determining one or more usable biometric regions within each of the one or more exposed biometric areas; andmodifying the one or more usable biometric regions while keeping aesthetics.2.The method [300] as claimed in claim 1, wherein the image includes at least one of a color impression or an exposed fingerprint of the one or more exposed biometric areas associated with the user.3.The method [300] as claimed in claim 1, wherein the one or more exposed biometric areas associated with the user in the image are identified based on segmenting the image to extract at least one of a hand, a fingerprint, and a fingertip edge.4.The method [300] as claimed in claim 3, wherein determining the one or more usable biometric regions within the one or more exposed biometric areas comprises:enhancing the segmented image to extract feature points, wherein each of the feature point comprises a plurality of minutiae points;clustering the plurality of minutiae points to obtain the one or more usable biometric regions, wherein each of the one or more usable biometric regions comprises a predefined number of minutiae points; andcalculating a usability score of each of the one or more usable biometric regions, based on a collective characteristic of each of the minutiae point from the plurality of minutiae points within each of the usable biometric region.5.The method [300] as claimed in claim 4, wherein the extracted feature points further comprise one or more user specific feature points comprising at least one of a mole and a scar.6.The method [300] as claimed in claim 4, wherein modifying the one or more usable biometric regions comprises:modifying each of the one or more usable biometric regions using a Feature Guided Fingerprint Generative (FGFG) module, based on the usability score of each of the one or more usable biometric regions.7.The method [300] as claimed in claim 4, wherein each of the minutiae points within each of the usable biometric region from the one or more usable biometric regions defines at least one of a ridge, a valley, a bifurcation, a dot, a ridge ending, and an island.8.The method [300] as claimed in claim 4, wherein the collective characteristic of each of the minutiae points comprises an assigned weight to at least one of a type of each minutiae point, a distance of each minutiae point from a core of a corresponding usable biometric region, and a local ridge density around each minutiae point.9.The method [300] as claimed in claim 8, wherein the core is a central reference point in the corresponding usable biometric region from the one or more usable biometric regions.10.The method [300] as claimed in claim 8, wherein the local ridge density around each minutiae point is a number of ridges present in the corresponding usable biometric region from the one or more usable biometric regions, around the core.11.The method [300] as claimed in claim 1, wherein keeping the aesthetics of the image comprises maintaining color, texture, art, paint and image applied on the one or more usable biometric regions.12.A system [100] for modifying a user biometric from an image, the system comprising:at least one processing unit [102];a memory unit [104] connected to the processing unit [102],wherein the processing unit [102] is configured to:identify one or more exposed biometric areas associated with the user in the image;determine one or more usable biometric regions within each of the one or more exposed biometric areas and;modify the one or more usable biometric regions while keeping aesthetics.13.The system as claimed in claim 12, wherein the image includes at least one of a color impression or an exposed fingerprint of the one or more exposed biometric areas associated with the user.14.The system as claimed in claim 12, wherein the one or more exposed biometric areas associated with the user in the image are identified based on segmenting the image to extract at least one of a hand, a fingerprint, and a fingertip edge.15.The system as claimed in claim 14, wherein the at least one processing unit [102] is further configured to:enhance the segmented image to extract feature points, wherein each of the feature point comprises a plurality of minutiae points;cluster the plurality of minutiae points to obtain the one or more usable biometric regions, wherein each of the one or more usable biometric regions comprises a predefined number of minutiae points; andcalculate a usability score of each of the one or more usable biometric regions, based on a collective characteristic of each of the minutiae point from the plurality of minutiae points within each of the usable biometric region.

Images