DESC:
FORM 2
THE PATENTS ACT, 1970
[39 of 1970]
&
THE PATENTS RULES, 2003
COMPLETE SPECIFICATION
(Section 10; Rule 13)

SYSTEM AND METHOD FOR PARTIAL FINGERPRINT ENROLLMENT AND MATCHING USING SMALL SIZE FINGERPRINT SENSORS

SAMSUNG R&D INSTITUTE INDIA – BANGALORE PRIVATE LIMITED,
#2870, Orion Building, Bagmane Constellation Business Park,
Outer Ring Road, Doddanekundi Circle,
Marathahalli Post, Bangalore – 560037,
Karnataka, India,
an Indian Company

The following Specification particularly describes the invention and the manner in which it is to be performed

RELATED APPLICATION

The present invention claims benefit of the Indian Provisional Application No. 332/CHE/2015 titled "ENROLLMENT METHODOLOGY FOR SMALL SIZE FINGERPRINT SENSORS” by Samsung R&D Institute India – Bangalore Private Limited, filed on 23rd January 2015, which is herein incorporated in its entirety by reference for all purposes.

FIELD OF THE INVENTION

The present invention generally relates to the field of biometric authentication, and more particularly relates to a system and method for fingerprint enrollment and matching using small size fingerprint sensors.

BACKGROUND OF THE INVENTION

Biometric authentication can be achieved by using any of the biometric data that can include, but not limited to, fingerprint, iris, face, retina, voice and the like. A typical biometric authentication system involves two main stages: enrollment and authentication. During enrollment, several biometric samples of a user are acquired and system registers the user by storing biometric information pertaining to the samples in the device’s internal database. At the time of authentication, a biometric sample of the user is captured again and its biometric information is matched against the one stored in the database.

For instance, during fingerprint enrollment, several fingerprint images of the user are acquired; useful features are extracted and stored in the device’s internal database. At the time of authentication, the user provides the fingerprint again. The system extracts features and matches them against the ones stored in the database. Although fingerprint enrollment is performed only once, it is a very crucial step as accuracy of authentication depends on the quality and completeness of enrolled samples.

Fingerprint authentication in current mobile devices is not limited to mere device unlocking but it is also used for high security applications such as mobile payments, securing sensitive information like health records, etc.

The use of embedded fingerprint sensors in mobile devices for authentication is prevalent. Typically, these sensors are miniaturized for cost and space constraints resulting in acquisition of partial fingerprints. Figure 1 is a schematic diagram 100 illustrating typical fingerprint sensor dimensions along with average fingerprint size, according to an existing art. According to the Figure 1, the size of an average fingerprint 102 is about 0.5” x 0.7”. But, the existing mobile devices primarily use rectangular shaped touch sensors 106 of size 0.45” x 0.2” or square shaped touch sensors 104 of size 0.2” x 0.2” approximately for fingerprint acquisition. Clearly, the platen area of these sensors is too small to capture entire finger thus requiring complex enrollment and matching methodology.

Figures 2a- 2b are schematic diagrams illustrating enrollment and authentication process using finger scanning in user equipment (UE) 202, according to an existing art. According to the Figure 2a, during enrollment process, a user scans his finger on a small scanner present on the UE 202. As the scanner present in the UE 202 is small, therefore plurality of images is scanned for the single finger. The plurality of scanned images are then combined to obtain a single fingerprint image using any of the known image generation techniques, such as, but not limited to, image stitching, and the like. The features extracted from the plurality of images are then stored as biometric information in an internal database 204 residing in UE.

Further, according to Figure 2b, when the user wishes to access any of the services from UE 202, then the user needs to be authenticated for accessing the service. During authentication, the same scanner from the UE 202 receives the fingerprint scan of the user’s finger. Upon obtaining the fingerprint scan, the features are first extracted and compared against the features pertaining to fingerprint scans obtained during enrollment process stored in the database 204 to authenticate the user before allowing access to the services.

Solid-state fingerprint sensors present on equipment, such as, but not limited to mobile devices, tablet devices, PDA, laptop, and the like for acquiring fingerprints can be of type touch or swipe. As shown in Figure 1, these sensors are miniaturized due to cost and space constraints which makes fingerprint enrollment and authentication on mobile devices complex procedures. In case of touch-based sensors, during enrollment, user is asked to provide multiple scans of the finger by lifting and touching the finger on the sensor in order to capture as much of finger portion possible. Although it is a one-time activity, it can be tedious and time-consuming. In case of swipe sensors, user needs to swipe finger multiple times during enrollment. Using overlapping partial scans, whole fingerprint is reconstructed using techniques like image stitching. This approach is proven to be underperforming since image stitching can be inaccurate and swiping finger on the sensor each time for authentication can be inconvenient especially for one-handed use of the mobile device.

Thus, there is a definite need for an improved method and system for faster, user-guided, and efficient biometric enrollment for mobile devices with small fingerprint sensors.

The above mentioned shortcomings, disadvantages and problems are addressed herein and which will be understood by reading and studying the following specification.

SUMMARY OF THE INVENTION

The various embodiments herein disclose a system and method for enabling fingerprint registration on a mobile device. The method receiving at least one first image of a fingerprint input in a first orientation, i.e. in a horizontal position, receiving at least one second image of the fingerprint input in a second orientation which is a vertical position wherein the first orientation and the second orientation are substantially perpendicular to each other. In an embodiment of the present invention, the first image of the fingerprint input can be provided in a vertical orientation and the second image of the fingerprint input can be provided in a horizontal orientation.

According to an embodiment of the present invention, the at least one first image and the at least one second image corresponds to partial horizontal finger scans and partial vertical finger scans. According to the present invention, the method further comprises of providing a visual indication to the user for rotating the mobile device based on an input from an orientation sensor in the mobile device.

According to the present invention, the method further comprises of estimating a finger coverage dynamically by stitching the at least one first image with the at least one second image and providing a visual indication of the estimated finger coverage on the device display. Further a class of the fingerprint is determined from the captured at least one first image for matching the fingerprint hierarchically.

According to an embodiment of the present invention, the method further comprises of using the fingerprint scans obtained during above-mentioned enrollment protocol for controlled authentication. The controlled authentication comprises of providing a visual guidance on a display of the mobile device to the user for placing one or more portions of the finger in one or more orientations, receiving, by the fingerprint sensor, input of one or more portions of the finger in one or more orientations, and authenticating the received one or more portions of the finger in the one or more orientations by comparing with the one or more portions of the registered fingerprint in the one or more orientations and validating the given fingerprint scan as per visual guidance provided.

Embodiments further disclose a mobile device comprising a fingerprint registration unit configured to receive at least one first image of a fingerprint input in a horizontal orientation, receive at least one second image of the fingerprint input in a vertical orientation, an orientation sensor for providing a visual indication to the user for rotating the mobile device, a pattern generation unit configured to estimate a finger coverage dynamically by stitching the at least one first image with the at least one second image, provide a visual indication of the estimated finger coverage, and determine a fingerprint class from the captured at least one first image for matching the fingerprint hierarchically, and a storage unit configured to store the predicted class pattern along with fingerprint features extracted for the enrolled fingerprint.

The foregoing has outlined, in general, the various aspects of the invention and is to serve as an aid to better understanding a more complete detailed description which is to follow. In reference to such, there is to be a clear understanding that the present invention is not limited to the method or application of use described and illustrated herein. It is intended that any other advantages and objects of the present invention that become apparent or obvious from the detailed description or illustrations contained herein are within the scope of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS

The other objects, features and advantages will occur to those skilled in the art from the following description of the preferred embodiment and the accompanying drawings in which:

Figure 1 is a schematic diagram illustrating fingerprint sensor dimensions embedded in mobile devices along with an average fingerprint size, according to an existing art.

Figures 2a-2b are schematic diagrams illustrating enrollment and authentication processes using fingerprint scanning, according to the existing art.

Figure 3a-3b is a schematic diagram comparing the conventional method of vertical finger scanning and horizontal finger scanning disclosed by the embodiments of the present invention.

Figure 4a-4c shows schematic diagrams illustrating a first stage of user guided enrollment process for obtaining partial fingerprints, according to an embodiment of the present invention.

Figure 5a- 5c are schematic diagrams illustrating obtaining fingerprints of the user by rotating the mobile device during enrollment process, according to an embodiment of the present invention.

Figure 6a- 6c are schematic diagrams illustrating second stage of user guided enrollment process for obtaining partial fingerprints, according to an embodiment of the present invention.

Figure 7a- 7b is a schematic diagram illustrating generating a fingerprint pattern using an image stitching of pre-scanned fingerprints, according to an embodiment of the present invention.

Figure 8 is a schematic diagram illustrating predicting the class of partial fingerprint using curve fitting, according to an embodiment of the present invention.

Figure 9 is a graphical representation of False Reject Rate against (FRR) against finger coverage at 0% false accept rate (FAR) according to an embodiment of the present invention.

Figure 10 is a schematic flow diagram illustrating a method of controlled authentication of the user through fingerprint scanning, according to an embodiment of the present invention.

Figures 11a and 11b are schematic diagrams illustrating controlled authentication of the user through fingerprint scanning, according to an embodiment of the present invention.

Figure 12 is a schematic flow diagram 1200 illustrating a method for enabling fingerprint registration on a mobile device, according to an embodiment of the present invention.

Figure 13a and 13b illustrate examples of horizontal and vertical partial scans acquired during stage 1 and stage 2 of the proposed enrollment process respectively.

Although specific features of the present invention are shown in some drawings and not in others. This is done for convenience only as each feature may be combined
with any or all of the other features in accordance with the present invention.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides a system and method for enabling fingerprint enrollment using small size fingerprint sensors embedded on the mobile devices. In the following detailed description of the embodiments of the invention, reference is made to the accompanying drawings that form a part hereof, and in which are shown by way of illustration specific embodiments in which the invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention, and it is to be understood that other embodiments may be utilized and that changes may be made without departing from the scope of the present invention. The following detailed description is, therefore, not to be taken in a limiting sense, and the scope of the present invention is defined only by the appended claims.

The specification may refer to “an”, “one” or “some” embodiment(s) in several locations. This does not necessarily imply that each such reference is to the same embodiment(s), or that the feature only applies to a single embodiment. Single features of different embodiments may also be combined to provide other embodiments.

As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless expressly stated otherwise. It will be further understood that the terms “includes”, “comprises”, “including” and/or “comprising” when used in this specification, specify the presence of stated features, integers, steps, operations, elements and/or components, but do not preclude the presence or addition of one or more other features integers, steps, operations, elements, components, and/or groups thereof. As used herein, the term “and/or” includes any and all combinations and arrangements of one or more of the associated listed items.

Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

The embodiments of the present invention, discloses a two-stage user guided enrollment protocol for mobile devices having rectangular fingerprint sensor.
The first stage of enrollment involves positioning the finger on the sensor longitudinally (along the length of finger) on the sensor. The rectangular shape of the fingerprint sensor, enables capturing of a wider finger area, thereby providing increased finger coverage within few scans.

The present invention provides a system and method for enabling fingerprint enrollment and matching using mobile devices having small size fingerprint sensor. Enrollment protocol involves placing the fingers vertically and horizontally over the fingerprint sensor thereby acquiring global ridge portion of the finger. Figure 3a-3b is a schematic diagram comparing the conventional method of vertical finger scanning and horizontal finger scanning method disclosed by the present invention. The horizontal or longitudinal scanning of the user finger ensures that the central portion of the finger is captured in the right manner. As the central portion of the finger is rich in biometric information and has high discriminating power, it is crucial for matching. The probability of capturing global ridge pattern in horizontal scans is high as compared to vertical scans.

According to an embodiment of the present invention, a method for enabling fingerprint registration on a mobile device comprises of receiving at least one first image of a fingerprint input in a first orientation. The mobile device herein can be any of the electronic devices with small size fingerprint sensor including, but not limited to, mobile phones, tablets, laptops, PDA, wearables, authentication devices at various high security zones such as banks, offices, and the like.

The mobile device comprises of a fingerprint enrollment unit that can receive the one or more images of the fingerprint input in the first orientation, i.e. horizontal. In an embodiment of the present invention, the fingerprint registration unit can be any of the known fingerprint sensors that can receive one or more first image of the fingerprint input in the first orientation. The person having ordinarily skilled in the art can understand that any of the known fingerprint sensors can be used as the fingerprint registration unit to capture one or more first images of the fingerprint input in the first orientation, without departing from the scope of the invention. Instead of placing finger horizontally on the fingerprint sensor as shown in Figure 4, the mobile device can be rotated as well such that finger is aligned longitudinally with the fingerprint sensor as shown in Figure 5.

Further, the method comprises steps of receiving at least one second image of the fingerprint input in a second orientation, wherein the first orientation and the second orientation are substantially perpendicular to each other. After receiving the one or more first image of the fingerprint input in the first orientation, the fingerprint enrollment unit can receive one or more second image of the fingerprint input in a second orientation. As shown in Figure 6, user is prompted to lift and move finger upwards in discrete steps in order to scan the complete finger in a vertical direction.

Further, the present invention also describes a controlled authentication utilizing fingerprint scans acquired using enrollment protocol for high security applications. The embodiments and use cases disclosed in the present invention are merely for clear understanding of the present method and device, but not to limit the scope of the present invention.

In another embodiment of the present invention, the first orientation for obtaining one or more first image of the fingerprint input can be vertical orientation of the user’s finger and the second orientation for obtaining one or more second image of the fingerprint input can be horizontal orientation of the user’s finger. In another embodiment of the present invention, the first orientation for obtaining one or more first image of the fingerprint input can be of horizontal orientation, and the second orientation for obtaining one or more second image of the fingerprint input can be of vertical orientation. The one or more images of the finger in both first and second orientation are captured as fingerprint inputs, to cover almost all the regions of the finger. Further, the one or more captured fingerprint images are checked for quality, and one or more bad quality images are discarded or replaced with good quality by re-capturing images for secure and efficient enrollment.

Further, the method comprises of providing, by an orientation sensor, a visual indication to the user for rotating the mobile device. After acquisition of one or more first images of the finger input, and identifying that the mobile device has captured sufficient fingerprint inputs in the first orientation, the mobile device informs the orientation sensor that fingerprint inputs are to be captured in the second orientation. Based on the received information, the orientation sensor provides visual guidance to the user to begin acquiring fingerprint scans in the second orientation by changing the orientation of the mobile device.

According to an embodiment of the present invention, the orientation sensor can be a gyroscopic sensor. In an embodiment of the present invention, the visual guidance for rotating the mobile device can be a video demo on the user interface (UI) or display of the mobile device. In another embodiment of the present invention, the orientation sensor can also provide audio guidance along with the video guidance for obtaining fingerprint input in the second orientation, without departing from the scope of the present invention.

In another embodiment of the present invention, the orientation sensor of the mobile device can provide visual guidance to the user for providing fingerprint input in the first orientation. Upon receiving the information from the mobile device, the orientation sensor can provide visual guidance for rotating the mobile device, and further provide visual guidance for fingerprint input in the second orientation, without departing from the scope of the present invention.

According to an embodiment of the present invention, the one or more first images and the one or more second images corresponds to partial horizontal and vertical finger scans. Further, the method comprises of estimating, by a pattern generation unit, a finger coverage dynamically by stitching the at least one first image with the at least one second image. Consecutive two or more first images may not have sufficient overlap for image stitching but it is likely they have good overlap with one or many second images. Upon obtaining one or more first images of the fingerprint in the first orientation and one or more second images of the fingerprint in the second orientation, the pattern generation unit can dynamically stitch the one or more first images and the second images and identify how much portion of the finger has been covered during fingerprint scanning with both the first orientation and the second orientation. Due to aid of second images, stitching of first images can be significantly improved and vice versa. The embodiments herein use, any of the conventional image stitching process to dynamically stitch the one or more fingerprint images to construct a final fingerprint pattern, without deviating from the scope of the invention.

Further, the method comprises of providing a visual indication of the estimated finger coverage. Upon stitching the one or more first images and the one or more second images, the pattern generation unit can provide the stitched image to the mobile device, wherein the mobile device can display the finger coverage obtained from stitched image on the user interface of the mobile device to guide the user dynamically.

The method further comprises of determining a fingerprint class to which the captured fingerprint belongs to from the captured at least one first image for matching the fingerprint hierarchically. The method comprises the step of storing a class for a registered fingerprint. In an embodiment of the present invention, the storage unit for storing the information about registered fingerprint can be any of the storage unit that includes, but not limited to, internal mobile memory, external storage device, server, third party database, and the like. Further, during authentication, the stored class information can be used for hierarchical matching.

The method as disclosed herein provides for full fingerprint coverage (~100%) in fewer scans compared to existing enrollment methodologies, thereby ensuring high authentication accuracy with very low False Reject Rate (FRR).

The enrollment protocol involves acquiring consecutive scans of the finger thereby knowing each scan corresponds to which part of finger. System can leverage this for high security applications for providing enhanced security. The system prompts the user to provide specific part/orientation of the finger using visual guidance. The user will be authenticated only if the required part/orientation of the finger is provided.
The orientation of the finger placed can be validated by calculating the relative rotation of the verify scan with respect to the genuine best match from the enrolled gallery.

The process of enrollment and authentication of the fingerprint based on the comparison with the enrolled fingerprint are described in detail based on the following diagrams.

Figure 3a-3b is a schematic diagram comparing the conventional method of vertical finger scanning and horizontal finger scanning disclosed by the embodiments of the present invention.`

According to Figure 3a, the fingerprint scanning for the enrollment of the fingerprint comprises of scanning of the user finger in a vertical orientation. During scanning in the vertical orientation 302, the user places his finger on fingerprint scanner vertically on the fingerprint scanner in the user mobile device. In 302, it can also be seen that, the mobile device can display video demo of how and which portion of the finger has to be placed on the fingerprint scanner so that the fingerprint image can be obtained optimally. Figure 3a also illustrates an example of a fingerprint scan, where a scanned image of a tip portion of the user finger in a vertical orientation.

According to Figure 3b, the fingerprint scanning for the enrollment of the fingerprint comprises of scanning of the user finger in a horizontal orientation 304. The user places his finger on fingerprint scanner on the mobile device horizontally. In 304, it can also be seen that, the mobile device can display video demo of how and which portion of the finger has to be placed on the fingerprint scanner so that the fingerprint image can be obtained in the horizontal orientation. In an example scan, it can be seen that the central ridge pattern of the finger of whorl class is captured. All the information related to the fingerprint scan is then saved in the storage unit for further authentication purposes.

Figure 4a-4c are schematic diagrams illustrating the first stage of user guided enrollment protocol for obtaining partial fingerprints, according to an embodiment of the present invention. According to the Figure 4, the various steps of receiving partial fingerprints in horizontal orientation are described. As shown in Figure 4a, at 402, the mobile device provides a visual indication, with a message “lift and tap finger from left to right” for instance, along with the video of the leftmost portion of the finger being placed on the fingerprint sensor. Upon placing the finger on the fingerprint sensor horizontally, the fingerprint sensor captures the image of the finger in a horizontal direction.

As shown in Figure 4b, at 404, the mobile device provides a visual indication, with a message “lift and tap finger from left to right” for instance, along with the video of the central portion of the fingertip being placed on the fingerprint sensor. The user then places the central portion of the fingertip on the fingerprint sensor as indicated by the demo video and corresponding part of the finger is captured.

As shown in Figure 4c, at 406, the mobile device displays a video demo with a message “lift and tap finger from left to right” along with the video of the right portion of the finger being placed on the fingerprint sensor. The user also places the right peripheral portion of the finger on the fingerprint sensor exactly based on the demo video. Similarly other video demos can be shown on the display of the mobile device and based on the finger placement; a plurality of partial fingerprints can be captured in horizontal direction from the left to right direction to scan the peripheral and central portions of the finger. Since horizontal scans provide more discrete information, for improving the finger coverage, finger scanning from left to right direction is performed at two locations starting with fingertip and central portion. Figure 13a shows the horizontal scans {H1…H5} corresponding to left to right finger scanning performed in 5 steps. The number of steps can vary with the sensor dimension for optimal finger coverage.

Figure 5a- 5c are schematic diagrams illustrating obtaining fingerprints of the user by rotating the mobile device during enrollment process, according to an embodiment of the present invention. The user places his finger on the fingerprint sensor in a vertical orientation and the mobile device continues receiving fingerprint input for horizontal orientation as well by rotating the mobile device. According to the present invention, a gyroscopic sensor present in the mobile device is used to identify the orientation of mobile and guide the user to rotate the device in order to obtain the first and second image sets.

According to Figure 5a, the mobile device acquires the fingerprint scan in vertical orientation at 502. Upon obtaining the fingerprint scans in the vertical orientation, the mobile device displays video demos to rotate the mobile device either in a left direction or right direction so that fingerprint scans can be obtained in the horizontal orientation as well.

According to Figure 5b, at 504, the gyroscopic sensor identifies that the mobile device is being rotated in the right direction based on the inclination and a tilt angle of the mobile device in the right direction. Upon detecting the direction and angle of tilt, the mobile device displays a video demo to keep rotating the mobile device in the right direction, for a next set of fingerprint images.

According to Figure 5c, at 506, the gyroscopic sensor identifies that the mobile has been rotated by 90 degrees in the right direction. The mobile device identifies that the user can enter fingerprints in a horizontal direction, and thus provides an indication to stop rotation of the mobile device. The indication can be a video demo played on the mobile device to indicate for stopping the rotation of the mobile device. Upon stopping the rotation, the fingerprint sensor can initiate capturing fingerprint images/scans in the horizontal direction. In another embodiment of the present invention, the mobile device can be held constant and the finger of the user can be rotated over the fingerprint sensor, such that fingerprint scans can be acquired in the horizontal orientation.

Figure 6 is a schematic diagram illustrating second stage of user guided enrollment process for obtaining partial fingerprints, according to another embodiment of the present invention. According to Figure 6, examples of one or more video demos showing the user how to capture the plurality of fingerprints in a vertical orientation of the user finger is described herein. As shown in Figure 6a, at 602, the mobile device displays a video demo with message “touch your finger tip on sensor”. The visual indication, can be, for instance a video of the fingertip being placed on the fingerprint sensor. Based on the guidance provided by demo the user places the fingertip on the fingerprint sensor and the corresponding fingerprint image is captured.

As shown by Figure 6b, at 604, the mobile device provides a visual indication, with a message “lift and move the finger upwards” for instance, along with the video of the lower portion of the fingertip being placed on the fingerprint sensor. The user then places the lower portion of the fingertip on the fingerprint sensor as indicated by the demo video and corresponding part of the finger is captured. As shown by Figure 6c, at 606, the mobile device displays a video demo with a message “lift and move the finger upwards” along with the video of the middle portion of the finger being placed on the fingerprint sensor. The user also places the middle portion of the finger on the fingerprint sensor exactly based on the demo video. Similarly other video demos can be shown on the display of the mobile device and based on the finger placement; a plurality of partial fingerprints can be captured. As shown in Figure 6c, user is prompted to lift and move finger upwards in few discrete steps in order to scan complete finger vertically. Figure 13b shows examples of partial vertical scans {V1…V5} obtained by scanning finger vertically starting with fingertip in 5 steps. The number of steps can vary with the sensor dimension for optimal finger coverage.

Figure 7a-7b is a schematic diagram illustrating generating a fingerprint pattern using an image stitching of pre-scanned fingerprints, according to an embodiment of the present invention. A plurality of both vertically scanned images and horizontally scanned images can be taken, which are then stitched together using at least one of a conventional image stitching method, to obtain a complete image corresponding to the fingerprint scan of the user.

According to the Figure 7a, two vertical scanned images, namely partial vertical scan 1(image 702) and partial vertical scan 2 (image 704) are considered for image stitching. However, these images 702 and 704 do not have sufficient overlap for enabling image stitching. Consecutively, one partial horizontal scan 1 image 706 and one partial vertical scan 1 image 702, obtained during fingerprint scanning in horizontal and vertical direction can be considered for image stitching. Two or more consecutive vertical images may not have sufficient overlap for image stitching but it is likely that they have sufficient overlap with one or many horizontal images. Due to the aid of horizontal images, stitching of vertical images can be significantly improved and vice versa.

First, the partial horizontal scan 1 706 and the partial vertical scan 1 702 can be stitched together to obtain a stitched image 1 708. Further, the stitched image 1 708 is stitched with the partial vertical scan 2 704 to obtain the stitched image 2 710. As shown in Figure 7b, the stitched image 2 710 has a larger finger portion containing richer biometric information useful for matching the fingerprint. According to the present invention, any of the known stitching processes can be used for stitching the plurality of partial scans to form the fingerprint image, without departing from the scope of the invention.

Figure 8 is a schematic diagram 800 illustrating predicting class of the partial fingerprint using curve fitting, according to an embodiment of the present invention. According to the diagram 800, the class of the fingerprint can be identified using partial horizontal scans. As the central ridge pattern of the finger is captured during horizontal scanning of the fingerprint, the curve-fitting like approaches can be used to predict the class of the fingerprint using the orientation map of underlying ridge flow. The information about the class of the fingerprint can be used in hierarchical matching. According to the present invention, the classes of the fingerprint can be at least one of, but not limited to, left loop, whorl, right loop and the like.

Further Figure 8 illustrates pre-defined kernels for Left loop (802), Whorl (804) and Right loop (806) of the user finger. Whenever horizontal partial fingerprint scans of the user are obtained, these three pre-defined kernels are fitted in the ridge orientation maps of the fingerprint scans using known kernel fitting techniques such as, but not limited to, on the ridge orientation maps and the like. Based on the response of kernel fitting, the fingerprint is classified into one of Left loop, Right loop or Whorl class. If none of the kernels fit well, the fingerprint class is categorized “unknown”.

According to the present invention, the fingerprint enrollment method and authentication system herein is used for various applications such as, but not limited to, conducting financial transactions, sharing confidential documents, defense applications, and the like which require secure authentication. The present invention of fingerprint enrollment method and matching ensures better finger coverage in fewer scans compared to existing enrollment methodologies, and thereby ensuring higher authentication accuracy with very low False Reject Rate (FRR).

Figure 9 is a graphical representation 900 of False Reject Rate (FRR) against finger coverage at 0% False Accept Rate (FAR), according to an embodiment of the present invention. The graph 900 is obtained by plotting finger coverage against False Rejection Rate (FRR) at 0% False Accept Rate (FAR). FRR indicates the rate at which the fingerprint authentication mechanism of the mobile can reject the fingerprint of the already enrolled user based on the obtained fingerprint. The graph 900 clearly shows that as the finger coverage is less, the chance of getting false rejects is higher. FAR indicates the rate at which the fingerprint authentication mechanism of the mobile can falsely accept unregistered finger as genuine.

Further, the present invention can be used in controlled authentication for high-security applications. The present invention enables acquiring consecutive scans of the finger, thereby knowing which scan corresponds to which part of finger. Further the user can be prompted to give specific part/orientation of the finger at the time of authentication by utilizing the mapping facilitated by the proposed enrollment protocol. The user can be provided with a visual demo depicting how to place the finger on the fingerprint sensor. Upon receiving the fingerprint scan, the user can be authenticated if and only if the required part/orientation of the finger is provided. The required part of the placed finger can be validated by finding which scan out of the enrollment scans the verify scan matches best. For example, the tip of the finger should match best with the first scan obtained during vertical scanning. The controlled authentication can also be performed based on the orientation of the finger. The required orientation of the finger placed can also be validated by calculating the relative rotation of the fingerprint scan with respect to the best match from the enrolled database.

According to the present invention, during authentication of the user, the mobile device necessitates the user to provide specific part of the finger in a particular orientation using visual guidance. Then, user will be authenticated only if the required part of the finger is provided in the particular orientation, which matches with the enrolled fingerprint of the user.

In total, the fingerprint of the user obtained comprises of a plurality of partial fingerprints which can be a part of the tip of the finger, middle of the finger and the bottom of the finger. During controlled authentication, the mobile device asks for one or more portions of the finger or one or more portions of the finger in one or more particular orientations. The mobile device shows the video demo on the display unit indicating which portion of the finger is to be positioned on the fingerprint sensor for scanning and in what particular orientation. Based on the displayed video demo, the user can place the particular portion of the finger in the particular direction. The fingerprint sensor can capture the scan of one or more portions of the fingers in the one or more particular orientations. Upon providing the particular portion of the finger or particular portion of the finger in the particular orientation, the mobile device can match the fingerprint with the enrolled fingerprint for authenticating the user and validate if user placed in the finger in right manner as indicated in video demo.

Figure 10 is a schematic flow diagram 1000 illustrating a method of controlled authentication, according to an embodiment of the present invention. The user is already enrolled for the biometric authentication by providing his/her fingerprints, where the extracted features are saved in the storage unit. The person having ordinarily skilled in the art can understand that the user has already been enrolled with the mobile device using the fingerprint based enrollment method as described herein above and thus not described herein again to avoid repetition.

According to the flow diagram 1000, at step 1002, user initiates accessing a high security application. In an embodiment of the present invention, the initiation of the high security application can refer to, but not limited to, launching of the application, logging into the application, and the like, without departing from the scope of the invention. At step 1004, the mobile device provides visual guidance to the user to place the finger on the fingerprint sensor in specific manner. The specific manner can refer to placing of the particular portion of the finger on the fingerprint sensor in a particular orientation.

At step 1006, the user places the finger on the fingerprint sensor in the manner as specified in the visual guidance. The fingerprint sensor receives the fingerprints of the user in the specified manner. At step 1008, the mobile device authenticates the fingerprint of the user for accessing the high security application, wherein the mobile device compares whether the received fingerprint matches with one of the enrolled fingerprints stored in the database or not. If the fingerprints do not match, then at step 1010, the mobile device rejects the user’s access to the high security application. In an embodiment of the present invention, upon rejecting the fingerprint, mobile device can perform various actions, such as but not limited to, providing predetermined number of attempts to provide correct fingerprint scan, requesting for security question, reporting to the security administrator about breach/ attempt of access by un-authorized person, and the like.

If received fingerprint matches with one of the enrolled fingerprints, then at step 1012, the mobile device further checks whether the fingerprint is obtained in the same manner as the fingerprint specified and displayed on the visual guidance. If fingerprint is obtained in the specified manner, the mobile device provides access to the high security application based on controlled authentication. If the fingerprint is not inputted in the specified manner, then at step 1010 the mobile device rejects the access to the high security application, and can take any of the herein above mentioned action, without departing from the scope of the invention.

Figures 11a and 11b are schematic diagrams illustrating controlled authentication of the user via fingerprint scanning, according to an embodiment of the present invention. Figure 11a is a schematic diagram 1100 illustrating controlled authentication of the user based on matching a specific portion of the finger with the enrolled specific portion of the user. According to the diagram 1100, the user is asked to place the specific portion of the finger on the fingerprint sensor for authentication. For example, the user is asked to place the tip or middle portion of the finger on the fingerprint sensor, and the like.
Figure 11b is a schematic diagram 1110 illustrating controlled authentication of the user based on matching specific portion of the finger in a particular orientation with the enrolled specific finger portion of the user. According to the diagram 1110, the user is asked to place the specific portion of the finger on the fingerprint sensor in a particular orientation for authentication. The orientation of the finger placed can be validated by calculating the relative rotation of the fingerprint scan with respect to the genuine best match from the enrolled database.

Figure 12 is a schematic flow diagram 1200 illustrating a method for enabling fingerprint registration on a mobile device, according to an embodiment of the present invention. According to the flow diagram 1200, at step 1202, the method comprises receiving at least one first image of a fingerprint input in a first orientation. According to the present invention, first image of the fingerprint input can be received in the first orientation by a fingerprint registration unit of an electronic device as described herein above, and thus not repeated herein again to avoid repetition.

Further, at step 1204, the method comprises step of receiving at least one second image of the fingerprint input in a second orientation. In an embodiment of the
present invention, the first orientation and second orientation of the fingerprint input can be interchanged, without departing from the scope of the invention. Further, at step 1206, the method further comprises step of stitching the at least one first image with the at least one second image to generate the fingerprint for registration, wherein the first orientation and the second orientation are substantially perpendicular to each other.

Figures 13a and 13b are the schematic diagrams 1300 and 1310 illustrating fingerprint scans obtained in vertical and horizontal orientations respectively, according to an embodiment of the present invention. According to Figure 13a, the schematic diagram 1300 illustrates plurality of fingerprints {V1,…,V7} taken while the finger of the user is placed on the fingerprint sensor in vertical orientation. Further, Figure 13b, the schematic diagram 1310 illustrates plurality of fingerprints {H1,..., H10} taken while the finger of the user is placed on the fingerprint sensor in horizontal orientation.

The present embodiments have been described with reference to specific example embodiments; it will be evident that various modifications and changes may be made to these embodiments without departing from the broader spirit and scope of the various embodiments. Furthermore, the various devices, modules, and the like described herein may be enabled and operated using hardware circuitry, for example, complementary metal oxide semiconductor based logic circuitry, firmware, software and/or any combination of hardware, firmware, and/or software embodied in a machine readable medium. For example, the various electrical structure and methods may be embodied using transistors, logic gates, and electrical circuits, such as application specific integrated circuit.

Although the embodiments herein are described with various specific embodiments, it will be obvious for a person skilled in the art to practice the invention with modifications. However, all such modifications are deemed to be within the scope of the claims. It is also to be understood that the following claims are intended to cover all of the generic and specific features of the embodiments described herein and all the statements of the scope of the embodiments which as a matter of language might be said to fall there between

,CLAIMS:
CLAIMS
We Claim:

1. A method for enabling fingerprint registration on a mobile device, comprising:
receiving at least one first image of a fingerprint input in a first orientation;
receiving at least one second image of the fingerprint input in a second orientation, and
stitching the at least one first image with the at least one second image to generate near complete fingerprint.
wherein the first orientation and the second orientation are substantially perpendicular to each other.

2. The method of claim 1, further comprising providing a visual indication to the user for rotating the mobile device based on an input from an orientation sensor in the mobile device.

3. The method of claim 1, further comprising:
estimating a finger coverage dynamically from the fingerprint generated; and
providing a visual indication of the estimated finger coverage.

4. The method of claim 1, further comprising determining a fingerprint class from the captured at least one first image for matching the fingerprint hierarchically.

5. The method of claim 4, wherein class of the fingerprint is at least one of left loop, whorl and right loop.

6. The method of claim 1, wherein the at least one first image and the at least one second image corresponds to partial horizontal finger scans and partial vertical finger scans.

7. The method of claim 1, further comprising using the fingerprint scans obtained by the registration for controlled authentication, wherein the controlled authentication comprising:
providing visual guidance on a display of the mobile device to the user for placing one or more portions of the finger in one or more orientations;
receiving, by the fingerprint sensor, input of one or more portions of the finger in one or more orientations; and
authenticating the received one or more portions of the finger in the one or more orientations by comparing with the one or more portions of the registered fingerprint in the one or more orientations and validating it against the visual guidance provided.

8. A mobile device comprising:
a fingerprint registration unit configured to:
receive at least one first image of a fingerprint input in a first orientation;
receive at least one second image of the fingerprint input in a second orientation;
stitching the at least one first image with the at least one second image to generate the near complete fingerprint;
an orientation sensor for providing a visual indication to the user for rotating at least one of finger or the mobile device;
a pattern generation unit configured to:
estimate a finger coverage dynamically from the generated fingerprint;
provide a visual indication of the estimated finger coverage; and
determine a fingerprint class from the captured at least one first image for matching the fingerprint hierarchically;
a storage unit configured to store a fingerprint class along with fingerprint features extracted for a registered fingerprint.
a fingerprint authentication unit for high security applications configured to:
provide visual guidance on a display of the mobile device to the user for placing one or more portions of the finger in one or more orientations;
receive by the fingerprint sensor, input of one or more portions of the finger in one or more orientations; and
authenticate the received one or more portions of the finger in the one or more orientations by comparing with the one or more portions of the registered fingerprint in the one or more orientations and validating it against the visual guidance provided.

Dated this the 20th day of January 2016
Signature

KEERTHI J S
Patent agent
Agent for the applicant