KR102376749B1 - Methods and systems of enrollment and authentication - Google Patents

Abstract

An authentication method for recognizing a candidate is disclosed. During authentication, a previously stored registration image is presented to the candidate on a display. The candidate is instructed to present a reproduced image of the same scene and/or object to the camera while holding the camera (eg, a mobile camera) without being supported in free space with respect to the gazed scene or object. As an alternative, the candidate can view the object without being supported in free space with respect to the camera. Using the camera, candidate images of the gaze scene or object are captured and presented together with previously stored registration images. The candidate aligns the candidate image with a previously stored registration image. Upon sorting, the candidate image is verified as the authentic image of the user, and the candidate is authenticated as a previously registered user.

Description

{Methods and systems of enrollment and authentication}

Field of the present invention relates to identity verification using a camera and display-based user registration and authentication method.

Many resources are being invested in developing reliable identity verification techniques to prevent fraudulent activity in e-commerce. Existing identity verification methods may rely on data the user knows, such as a password, personal identification number (PIN), or other information, or an item the user possesses, such as a smart card.

Based on physiological and/or behavioral characteristics, considerable effort has been devoted to identity verification using biometrics. Characteristics used for biometric authentication may include fingerprint, retina or iris scan, voice, signature, and/or face.

U.S. Patent No. 7,327,858 discloses biometric authentication using the main lines of the hand when scanned by a scanning unit such as a scanner having a planar platen.

Various computerized biometric methods for registration and authentication of users are provided herein. During authentication, a previously stored registration image is presented to the candidate on the display.

The use of the term body part in the present invention is merely an example, and will be referred to as a “view object” defined later.

The user is instructed to present the body part to the camera while the user holds the body part without being supported in free space with respect to the camera. Also, the user can use the mobile camera to capture any fixed object/scene, and then is asked to align the camera with the object. Using the camera, candidate images of the body parts are captured and presented together with previously stored registration images. The candidate aligns the candidate image with a previously stored registration image. Upon sorting, the candidate image is verified as an authentic image of the candidate, and the candidate is verified as a previously registered person. A previously stored registration image may be generated independently of the candidate. The same registration image may be displayed for registration of a plurality of previously unregistered users.

During registration, the user can view the body part without being supported in free space with respect to the camera. Alternatively, the user can view the camera without being supported in free space against a fixed scene/object. The camera can be used to capture background and registration images of body parts. Authentication may include capturing of candidate images in the background and time series of the candidate body part.

During capture, the candidate images in the time series are presented on the display to be superimposed on the registration image with their respective backgrounds.

The view object may be an entire face, a hand, an ear, an eye, and a portion of an item such as a photograph or document. The candidate may align a feature of the object of view in the candidate image with a corresponding feature in the enrollment image. The features and corresponding features can be lines, ridges, petticoats, scars, moles, areas of skin pigmentation difference, birthmarks, and wrinkles. Moreover, features and corresponding features may be dots, edges, intrinsic visual traits, colors or textures in any item or region with a contrast that may be differentiated from the rest of the captured image of the object portion.

Verification may be performed by comparing the registration image with the authentication image. Alignment may include alignment of horizontal and vertical image coordinates, angular orientation in image space, and actual spatial distance to the camera.

Prior to presentation of the previously stored registration image, the registration image may be processed by performing a horizontal reflector inversion of the registration image.

The candidate may be instructed to align a feature of the object of view in the candidate image with a corresponding feature in the enrollment image. When the candidate aligns a feature of the object of view in the candidate image with a corresponding feature in the registration image, one or more of the candidate images are verified as the candidate's authentic image. Accordingly, one or more candidate images may be authenticated as the candidate's authentic images. The candidate is then authenticated as a previously registered user by comparing features in the candidate image to corresponding features in the enrollment image.

A pre-registered image of the viewing object may be captured when the user is holding the user's viewing object without being supported in free space with respect to the camera. A pre-registered image may be presented to the user on the display. Presenting the pre-registered image, the user can superimpose the symbol presented on the display with the imaged features of the object of view. By superimposing the symbol with the imaged features of the object of view, the registration image is stored along with the symbol's respective position and orientation in image space in the pre-registered image.

Symbols may emerge from a predefined group of symbols, such as symbols or symbols created by keyboard strokes. Alternatively, the symbol may be formed manually, such as a handwritten symbol, shape, or other symbol formed by a candidate.

During listing, the user can form and set up a predetermined sequence of steps, each step consisting of the appearance of a different/same symbol on the screen without capturing the view object. The detailed rules regarding symbols are referred to as one or more password symbols. The mentioned sequence steps may be further ordered such that only the candidate repeats the synchronized listing sequence steps known to the candidate during authentication.

Moreover, registration and authentication may include presenting the object of view within a predefined specific time frame during registration.

The time synchronization method may be implemented by any indication of time progress, such as a timer, progress bar, vibration, or other form of visual and audio indication of the length of time measured.

Another implementation may be any keyboard stroke over a time interval, with a visual symbol indicating the length of time of keyboard presses by the candidate, including gaps between the step sequences defined herein, established during registration.

The step sequence defined here is valid only when the candidate can regenerate or synchronize the same step sequence as at the time of entry in the authentication step. The present invention describes hints on the screen to help candidates reproduce the same registration step in the authentication step.

Prior to overlapping, the symbol may be repositioned, resized, or rotated. Presenting the registration image to the candidate on the display may include at least a portion of the image of the object of view and the stored symbol. Alternatively, presenting the registration image on the candidate's display may include presentation of the stored symbol without any portion of the image of the object of view.

A plurality of previously defined registration scales may be presented to the user on the display. The user may select one or more of the scales presented on the display for generating the selected registration scale. The user may overlay and/or rearrange one of the registration images with the selected scale. The selected scale may be saved as a registered image. Authentication includes presenting on a display a time series of candidate images superimposed on the selected authentication scale. The candidate may also align the candidate image with the selected registration scale.

During authentication, presentation of the registration scale to the candidate may include a plurality of registration scales, and the candidate may select the registration scale.

Provided herein are various mobile computer systems that operate to perform the computerized biometric methods disclosed herein, wherein a camera and a display are part of the mobile computer system.

Various computer-readable media encoded with processing instructions that cause a processor to execute the methods disclosed herein are provided herein.

BRIEF DESCRIPTION OF THE DRAWINGS The above and/or other aspects will become apparent from the following detailed description when considered in conjunction with the accompanying drawings.

1 shows a simplified block diagram of a mobile computer system for operation of different aspects of the present invention;
2 shows a flowchart of a method, according to an aspect of the present invention.
3 shows a registration image presented on a display with a background captured also in accordance with features of the present invention;
FIG. 4 shows registration and authentication images overlaid with their respective backgrounds, according to the method shown in FIG. 2 .
5 shows another biometric registration and authentication method according to an aspect of the present invention.
6 and 7 respectively show examples of registered images previously stored on a display, in accordance with an aspect of the present invention.
8 shows a flow diagram of a method, including different features of the present invention.
9 , 10a and 10b show a flow diagram of a method and a registration image, respectively, according to different aspects of the present invention.
11 shows an example of a registration and authentication method using time-series steps according to the method shown in FIGS. 5 and 9 .
Fig. 12 shows a diagram of a scaled symbol presented on a screen according to the method shown in Fig. 11;
13 , 14 , 15 show diagrams of the user possibility of superimposing an object with the scaled symbol of FIG. 12 according to the method shown in FIG. 11 .
15 shows an example of a registration and authentication method using time-series steps according to the method shown in FIGS. 5 and 9 .
16 shows an example of registration of hand-drawn symbols on the screen rather than other auxiliary symbols to display and/or hide registered symbols.

As used herein, the term “registration” refers to the spatial alignment of image features of different images captured at different locations and/or times of a similar or the same object within different image frames.

As used herein, the term “identification” refers to the receipt of initial information, name, identity card number, or other information to uniquely identify a candidate presented by a registered user prior to or during registration.

As used herein, the term “registration” refers to the initiation of a user within a biometric system while receiving and storing biometric information for subsequent identity verification or authentication. The term “registered image” refers to image capture during registration.

As used herein, the term "authentication" refers to biometric verification that a candidate is himself or herself during identification.

As used herein, the term “verification” refers to authentication or denial of authentication based on biometric image characteristics of a candidate as either a previously registered user or a previously identified person.

The term "superimpose" as used herein in the category of simultaneously presenting two or more images on a single display means (i) combining two or more images using image processing techniques and presenting the combined image, and and/or (ii) alternately presenting two or more images to the candidate such that the appearance to the candidate is that of the combined superimposed image.

The term “body part” as used herein for use in biometric enrollment and authentication may be a user's hands, eyes, ears, nose, facial contours, or other parts of a user.

The term “view image” as defined herein includes any reproducible visual scene captured by a camera or other imaging device. A scene may be movable or stationary (such as a door, building background, etc.) It can be composed of any type of object.

The term “password symbol” shall refer to either a signature, an image, any shape, symbol, or any combination of connected dots or pixels, any other form of symbol that may be used for identification, or any combination of these features. can As is known to those skilled in the art of image processing and biometrics, in authentication, a handwritten symbol will be authenticated as a biometric signature.

As used herein, the term “region of interest” refers to any portion of the view image that is selected and marked on the screen by a user during enrollment. In view image authentication, the hot spot portion should be independently correlated to the view image. Correlation processes are known to those skilled in the image processing art.

The term "viewed object" relates to capturing a view image of an object or scene that can be visually reproduced and that can be captured during registration, on- with candidate images of the same object and/or scene. It relates to capturing a view image of an object or scene that is repeatedly superimposed on a screen part/distorted image.

The term "feature" as used herein with respect to the alignment of a body part excludes the "contour" or outer edge of the body part.

As used herein, the term “user” refers to an end user, system user, and/or customer with reference to a general reference or registration phase of a person.

As used herein, “candidate” refers to an end user, system user, and/or customer involved in the authentication phase.

As used herein, a "time synchronization method" is a method of time synchronization to a user or candidate being utilized by any indication of time progress, such as a timer, progress bar, vibration, or other form of visual and audible indication of a measured length of time. It is related to the measurement of time by

As used herein, the term “time frame” relates to any measured duration that can be measured and reproduced by a user.

As used herein, the term “time-order steps” relates to steps that a candidate must follow in the same order and/or in time synchronization as recorded and presented by a user in the registration phase. These sequence steps may consist of on-screen superimposed candidate images of the same object or view. Moreover, these sequential steps may be any combination of symbols, on-screen placement, drawing, keyboard strokes, or any other combination thereof, and the like. One sequence step may present a view object for the duration of the measurement. The time steps may be a combination of predetermined durations for applying a step or a combination of durations between a series of steps.

As used herein, "one", "one", such as "one candidate image", "one registered image", "one background", "one or more candidate images", "one or more registered images" and "one or more backgrounds" has the meaning of "one or more".

The object must have an edge of some sufficient contrast and must be reproducible to allow the authentication step to succeed.

To introduce, different embodiments of the present invention find use in a number of different biometric methods for identifying and registering a user in a system and, subsequently, for authenticating a candidate as one of previously registered users. can do. Embodiments of the present invention are configured for use with two-dimensional or three-dimensional displays and cameras available in mobile computer systems, such as smart-phones. In some embodiments of the present invention, during registration, the user presents the viewing object (eg, hand) unsupported in free space, and the registration image is captured. Subsequently, for example, when the user wants to make an online transaction, for example, presenting a same-view object, such as a hand, and an authentication image captures the unsupported hand along with the free space between the camera and the hand as well. do. The authentication image may be treated as being the same hand used for the enrollment image, and authentication is realized, or authentication is denied otherwise.

When a registered image is used again for authentication, the registered image may be blurred when presented for authentication use, grayscale or color may be manipulated, resolution may be degraded, partially displayed or otherwise image processed. , it may be difficult for a crook to reconstruct the authentication image from the registered image when presented during authentication.

In various embodiments of the invention, which will be presented in further detail, authentication may take place using a different computer system than enrollment or at a different location and/or time or date, and for biometric enrollment and authentication, any viewing object may be may be appropriate.

The authentication process and identity verification may be similar to the processing techniques disclosed in US 7,327,858, such as comparing specific features between enrollment and authentication images. The features used in US 7,327,858 are the striking lines of the hand. However, according to the prior art method, image registration is performed before actual authentication processing is performed, during which respective images of the same features of interest are located in both the registration image and the authentication image.

Various embodiments of the present invention are configured to avoid or facilitate image registration in registration images and between common imaged features of interest in authentication. By avoiding or facilitating image listing, it is possible to reduce the computational load of the processor used and/or reduce the required computational time.

Thus, useful for avoiding and/or facilitating image entry between an enrolled image and an authenticated image to perform authentication using an unsupported view object placed in free space, or image capture of a mobile device that remains unsupported in relation to the object. It would be necessary and advantageous to have a method and system.

Although mobile computer systems are used for registration and authentication in some embodiments of the present invention, in other embodiments, non-mobile computer systems and otherwise dedicated computer systems may also or alternatively be used.

The registration and authentication process described herein must be combined from the aforementioned set of sequential steps. The sequence steps presented herein are candidate steps that must follow the same sequence and/or time synchronization. These sequencing steps may consist of on-script overlapping candidate images of the same object or view. Moreover, these sequence steps may be combinations of symbols on screen layout, drawings, keyboard strokes, or the like.

Before describing in detail embodiments of the invention, the invention is not limited in application to the details of design and arrangement of components illustrated in the drawings or set forth in the following description. The invention is not capable of other embodiments being practiced or carried out in various ways. Furthermore, the phraseology and terminology used herein should not be regarded as limiting.

Referring now to the drawings, FIG. 1 illustrates a simplified block diagram of a mobile computer system 100 in accordance with aspects of the present invention. Mobile computer system 100 is connectable to server 108 via network 106 . Mobile computer system 100 is also connectable to cellular network 121 via cellular base station transceiver 119 . Mobile computer system 100 includes a processor HOC coupled to local data storage 102 . A data communication module 104C couples the processor HOC to the data network 106 . The cellular communication module connects the processor HOC to the cellular network 121 . The mobile computer system 100 includes a display 109 coupled to the processor HOC, a global positioning system 107 , a camera 105 , a microphone 111 , a speaker 113 , a vibrator 115 , and an accelerometer/gravity sensor. , Bluetooth™, and HOC peripheral accessory devices such as IR sensors (not shown).

Reference is now made to FIG. 2 , which shows a method 301 for biometric enrollment and authentication in accordance with an aspect of the present invention. Method 301 includes user registration steps 303 , 305 , and authentication/non-authentication- or verification steps 307-317 of the candidate. At step 303 , an image of the object of view (eg, the candidate's hand) is captured, for example using the camera 105 of the mobile computer system 100 , for registration image capture of the body part.

Figure 2a is now shown, which shows a user holding a hand-like view image in free space with respect to camera 105 and/or display 109 (of IP AD™ used as an example of mobile computer system 100). See also. The display 109 shows the background including the user's face and torso, and the registered image 20 of the user's height, palm, wrist plus arm. The camera 105 and/or the display 109 may be mounted on a stand so that both hands of the user are free. FIG. 2B shows, by way of example only, other viewing objects, such as keys, cards, documents, ears or eyes, that may be captured to provide an enrollment image 20 .

Reference is now made to FIG. 3 , which shows, for example, a registration image 20 including only the key presented, on the display 109 of the mobile computer system 100 . One feature in accordance with an embodiment of the present invention includes avoidance of processing to find the outline of the registration image 20 of the object of view (eg, key). Therefore, in FIG. 3 , an image 20 of the view object key is presented along with the background captured during registration. The background is schematically shown with the image 22 of the bush of FIG. 4 .

2 , at step 305 , a registered image 20 is associated with the user. The association of the registration image 20 with the user may include the user's name, date of birth, identification (ID) number or social security or national insurance or social security number.

Step 307 initiates the authentication/non-authentication process of a previously registered candidate. The authentication may take place using a different computer system than the registration steps 303-307 or at a different location and/or time or date.

Reference is now made to FIG. 4 which shows the display 109 during authentication. In step 307 , registration image 20 is presented on display 109 along with background image 22 .

2 , in step 309 , the candidate is instructed to present an object of view (eg, key) to the camera 105 , and a time series of candidate images 40 , such as a video, is captured for the candidate's height. do. The candidate image 40 may include an image of the candidate's candidate view object and a respective background image 42 . According to one aspect of the invention, in the candidate image 20 , the candidate image 40 is a contour of the image of the object of view (eg key) to extract and present an image of the object of view (eg key without background). Each is presented with their own background, avoiding looking for them. In FIG. 4 , the background of a candidate image 40 of a hand is schematically shown together with an image 42 of a house.

Referring again to FIG. 2 , in step 311 , a time series of candidate images 40 is displayed on display 109 and captured together with previously captured registration images 20 in step 303 (step 303 ). 309)). Accordingly, in step 311 , one or more candidate images 40 are superimposed on the registration image 20 along with their respective backgrounds 42 , 22 .

At step 313 , the candidate aligns (i.e., places and relocate).

In order to realize the alignment in step 313 , after being captured (step 303 ) the registered image 20 and/or the candidate image 40 are horizontal before display (step 311 ) on the display 109 . It can be processed using horizontal mirror inversion or the like.

The basis for the alignment in step 313 may be a feature of the object of view (eg, key) in the candidate image 40 along with the corresponding feature in the registration image 20 . Such features and corresponding features may include lines, ridges, a line, scar, area of skin pigmentation difference, birth mark, or wrinkle of each body part or candidate body part, respectively. can be Additionally, this feature may be a contrast-based feature, such as an edge, line, point, or any visual differential region located within an object image object. Alignment step 313 includes alignment of horizontal and vertical image coordinates with respect to camera 105 , angular orientation in image space, and actual spatial distance to camera 105 .

At decision block 315 , a verification step 317 may be performed when the one or more candidate images 40 are aligned with the enrollment image 20 on the display 109 . During verification (step 317 ), the candidate compares the registered image 20 with a previously stored authenticated image according to any technique known in the image processing art, whereby the previously registered user and/or in steps 303-305 Alternatively, it may be verified as a previously identified user.

At decision block 315 , the alignment continues at step 313 by the candidate when one or more candidate images 40 are not yet aligned with the registration image 20 on the display 109 .

Reference is now made to FIG. 5 , which shows another biometric registration and authentication method 501 , in accordance with an aspect of the present invention. The biometric method 501 includes steps for registration and authentication of a user. At step 503 , the user's object of view may be captured using the camera 105 to create a registered image 20 of the object of view (eg, key). In step 305 the registered image 20 is associated with the user. Step 507 initiates the authentication phase, wherein the registration image 20 is presented on the display 109 with or without background 22 . At step 509 , the candidate is instructed to present the key to the camera 105 , and a time series of candidate images 40 is captured for the object of view (eg, the candidate's height). Candidate image 40 includes an image of a candidate view object of the candidate being presented, with or without background image 42 .

In step 511, the time series of candidate images is displayed (with or without background 22) during capture (step 509) with display of registered image 20 previously captured in step 503 (step 509). 109) is displayed. In step 511 , the candidate image 40 is superimposed on the registration image 20 .

At step 513 , the candidate may align, position, and reposition his or her hands and/or keys in front of the camera 105 such that the candidate image 40 is aligned with the enrollment image 20 on the display 109 . .

To facilitate alignment, after capture (step 503 ) registration image 20 and/or candidate image 40 may be processed using horizontal mirror inversion before being displayed on display 109 . there is.

The basis of the alignment in step 513 may be an image of a feature or height of a body part in the candidate image 40 along with a corresponding image of a feature in the registration image 20 . The features and corresponding features may be ridges, lines, scars, areas of skin pigmentation differences, birth marks, or wrinkles on a body part or height candidate body part or height. The alignment of step 513 may be performed without alignment of the imaged contour of the object of view (eg, key). The image outline is not necessarily visible on the display 109 for the candidate image 40 and/or the registration image 20 . Alignment step 513 of image features includes horizontal and vertical image coordinate alignment, angular orientation in image space, and actual spatial distance to camera 105 .

At decision block 315 , when the one or more candidate images 40 are aligned with the registered image 20 , authentication/disapproval may be performed. Authenticating the candidate as a previously registered user may be realized by comparing the registered image 20 to the authentication image. Authentication is verified by associating the candidate image with the candidate's authentication image (step 317).

When the time series of the candidate image 40 is not perfectly aligned with the registered image 40 on the display 109 in the decision block 315, the alignment of the candidate image 40 with the registered image 20 by the candidate is 513) continues.

Reference is now made to FIG. 6 , which shows a method 801 according to an aspect of the present invention. In step 807, the previously stored registration image is shown to the user for registration purposes. Reference is now also made to FIGS. 7 and 8 , which show examples of registration images 65 , 75 respectively previously stored on the display 109 . The registration image 65 is an ellipse, in which the user aligns the image of his or her hand 33 within the ellipse 65 . A dashed line 37 through the ellipse 56 may be used to align the images of the middle finger and wrist. Image 75 includes dashed lines to align respective images of the wrist and fingers of hand 33 .

Referring back to FIG. 6 , at step 809 , the candidate is instructed to present a view object (eg, height) or body part (eg, hand 33 ) to camera 105 . In step 809 , a candidate image 40 of the candidate is captured, and the candidate attempts to align the candidate image to a previously stored registration image displayed on the display 109 . With respect to image 65 and step 809 , the candidate aligns the image of each fingertip immediately within the perimeter of the ellipse 65 and the middle finger aligns symmetrically around the dashed line 37 passing through the ellipse 65 . do. With respect to image 75 and step 809, the candidate has the dashed line in image 75 placed in the center of each finger of his/her hand 33 and each finger of his/her hand 33 is located in the image ( 75) and try to ensure that it is parallel to the dashed line. In decision block 815, if the alignment of step 809 is realized, the verification is performed by verifying the candidate image 40 with the authenticated image of the registered user (step 317). Otherwise, the alignment by the candidate Continue to step 809 .

When a previously stored registration image 65/75 is used, the registered user, for example 10 stored in the memory 102 of the mobile computer system 100, to use for performing the registration capture step 503, for example. One or more previously stored registration images 65/75 may be selected from a bank of between 100 and 100 stored images. The selection of a previously stored registration image 65/75 provides additional security and allows the user to select a registration image 65/75 suitable for their hand 33 in this example.

Reference is made to FIGS. 9 and 10 , which show a method 901 and a registration image 95 respectively shown on a display 109 , in accordance with an aspect of the present invention. In step 903 , a registered image 95 of the user's body part is captured using the camera 105 . The body part shown in the registration image 95 (in FIG. 10A ) is the user's hand. The registration image 95 captured in step 903 is then presented to the user on a display 109 (step 905). On the registration image, the user is presented with a symbol over a body part feature presented on the display 109. 101 (eg, a rectangle) may be superimposed (step 907 ) The features of the hand shown in Fig. 10A may include ridges, lines, wrinkles, scars, or curved portions connecting the fingers. Superposition (step 907 ) may include the user repositioning, resizing, and/or rotating the symbols over the features.

At decision block 909 , upon successful superimposition of the symbols 101 with the hand features on the display 109 , the symbols 101 represent their respective positions, orientations, and hand and and/or stored with the features of symbol 101 (step 911). If the symbol 101 on the display 109 has not yet overlapped with the hand feature, then the overlap continues at step 907 .

Referring back to FIG. 9 , authentication of the candidate continues with steps 913 - 921 . In step 913 , the registration image 95 stored in step 911 is presented on the display 109 . In step 913 the mobile computer system 100 instructs the candidate to present the candidate body part to the camera. If the candidate body part is, for example, a hand, a time series of candidate images is captured for the hand using the camera 105 . During the capture step 913 , the time series of candidate images is overlaid with the registration image 95 (step 915 ) and displayed on the display 109 . At step 917 , at least one of the candidate images is aligned with the registration image 95 , such that the registration image 95 and corresponding features of one or more of the candidate images are aligned. Upon alignment (step 917 ), the candidate may move a hand in front of the camera 105 such that the identification features (eg lines, wrinkles, etc.) known only to the registered user are aligned with the symbol 101 . . At decision block 919 , upon successful alignment of the feature with the symbol, a verification 317 may then be performed, eg, based on the description of US Pat. No. 7,327,858. Otherwise, upon failure of alignment or denial of authentication, the candidate may again be instructed to present his or her hand again in front of the camera 105 using step 913 . Candidates may use the front or rear or other cameras of the mobile device simultaneously, or in a specific order.

Reference is now made to FIG. 10B , which shows another example of the use of symbols 101 , shown as software pegs 101 on display 109 . The software peg 101b may be placed by the user, for example, to determine the location of the curved connector where the software peck 101 connects the fingers.

Reference is now made to FIG. 11 , which includes a flow diagram illustrating a method 1101 in accordance with an aspect of the present invention. At step 1103 , the user is presented on display 109 with a plurality of previously defined registration scales. An example of a previously defined registration scale shown on display 109 is shown in FIG. 12 , as a scale or grid 1201a-1201c with respective gridlines 1203a-1203c.

Referring back to the method 1101 of FIG. 11 , at step 1105 , a registered image of the user's object of view is captured using the camera 105 . At step 1107 , a registered image of the user is presented to the user on display 109 . The user can select a grid scale 1201a-1201c (step 1109), and the registered image is overlaid with the selected grid or scale 1201a-1201c. By using the grid 1201b as the selected scale, the user can rearrange or rotate the selected grid scale 1201b so that the grid line 1203b is located under the women's shoe as shown in FIG. 14 , and the upper part of the door is shown in FIG. 14 . It is arranged along line 1203c as shown in Fig. Alternatively, the user rotates and positions the camera so that grid lines 1203b are positioned on three or four 1205b as shown in FIG. 14 . Referring back to method 1101 of FIG. 11 , in step 11110 a registered image 20 of the user's object of view is stored along with the selected grid scale 1201b and associated with the registered user.

The authentication steps 1113-1123 are steps for verifying whether or not the candidate is a user previously registered in steps 1101-1111. At step 1113 , an identification (ID) number of the candidate may be received by the mobile computer system 100 .

The candidate is instructed by the system 100 to present the object of view to the camera 105 to capture a candidate image 40 of the candidate object of view (step 1115 ).

The captured candidate images are presented (step 1117) and superimposed on the selected registration scale 1201. In step 1119, the candidate aligns one of the candidate images with the selected registration scale 1201. Decision block At 11210, if there is an alignment between the candidate image 40 and the selected scale 1201, the candidate image 40 may or may not be verified as the authentic image of the candidate as a previously registered user in step 1123. Otherwise, at decision block 11210, the alignment by the candidate continues at step 1119, or, after several alignment failures, the candidate is notified that authentication has failed.

Additional security features are involved in steps 1115 and 1117 . In order to verify the candidate image 40 as an authentic image (step 317 ), the candidate view object and features need to be identical to the view object and features used during registration. Moreover, the selected scale 1201 should be chosen by the candidate to be used during registration to verify that the candidate image is authentic.

Among the registration processes shown in FIGS. 13 and 14 , the mobile computer system 100 may not know the details (size, shape, etc.) of the user's view object to be registered. Accordingly, in the registration step, several grid scales 1201 with respective grid lines 1203 are displayed on the display 109 , and the user aligns his/her view object to each scale 1201 . The viewing object may be aligned to the scale 1201 , in which case the entire viewing object should be placed inside a rectangular box of the scale 1201 .

Referring to FIG. 13 , when the user aligns the view object door on the display 109 to each scale 1201 during registration, the result is that the object of view moves up to the camera 105 in the case of each scale 1201 . They may be actually arranged at different distances from each other. The mobile computer system 100 may select the best scale 1201 for the candidate, and features extracted from the registration image 20 related to the corresponding scale 1201 are the most robust and distinct. From here, the best selected scale 1201 can be used for the user, and the registered image can be stored and used during verification.

With respect to one of the many validation processes described above, the same registration mechanism can be used by one of the validation processes described above, where the candidate selects the required scale 1201 to assign his/her view object to the selection of object placement. should be placed accordingly. If too many verifications of the candidate are attempted, the candidate can alternatively align the object exactly the same as in the registration step without presenting information about the identity. The number of verification attempts of a candidate can now be limited to only relevant registrants of the same scale and rotation, and the candidate can be identified.

Using the grid 1201b as the selected scale, the user can rearrange or rotate the selected grid scale 1201b so that the grid lines are on the line 1205b at the top of the user's door as shown in FIG. 14 . . Alternatively, the user rotates and positions his or her hand so that the grid line 1203b is on the line 1205b of the top line of the user's door as shown in FIG. 15 . The alternatives presented by FIGS. 14 and 15 may apply to both the registration and verification process or only the registration process. In the verification process, the scale 1201 may be predefined and immutable on the display 109 .

The verification process may be repeated in a particular way. For example, during registration, the user selects one of the scales 1201a, 1201b, and 1201c to align his object 1205a and women's shoes to the line 1203b of the scale 1201b. If the verification is successful, the candidate continues the second verification step at another scale, and so on. For a more secure option, during registration the user can combine the scales 1201 with sequential verification steps, generating a password from the ordered order of the scales 1201 .

In another embodiment, a candidate user may capture a card and use a series of card placements on the screen and alignment with different scales or symbols. Card alignment steps and timing of card alignment with scale are registered so that fraudsters cannot extract the correct order and timing during authentication. In this way, the scammer will need a mobile device with the cards, a specific scale and placement of the cards in the right order, and the timing of each step. For example, a candidate aligns a card with a first scale for 2 seconds, then aligns a card on two or more scales for 0.5 second. The system will verify the authenticity and correct alignment of the cards at the card presentation timing and the registered sequence of scaled symbols.

The use of different cards of different sizes, such as driver's licenses and passports, requires different symbol sizes to ensure that the card is placed at the correct focal distance from the camera so that the card is in optimal focus. For example, a driver's license would have a small bounding rectangle on the screen, and a passport would require a much larger rectangle on the display. When a candidate wishes to authenticate a card, the card must align with the matching bounding rectangle on the screen. In this way, the card can be read at an optimal focal length, and card reading will not require an auxiliary device.

If an assistive device is desired, the mobile device is positioned such that the reflector reflects the inner palm of the hand. In this implementation, the mobile camera may capture the palm of the candidate and perform authentication. Due to the reflector, the candidate can see the display of the mobile device (due to the angle of the mobile placement) and interact with the mobile and front camera. Additionally, the user can capture the back of the hand and then place the hand on the bottom of the device. This enables interaction with the mobile device display while simultaneously identifying the candidate using the hand. In this particular implementation, the mobile device may be positioned by the candidate on the assistive device and is a photograph of a hand placed on a surface-type or peck-type device to stabilize the hand at a fixed location and distance from the mobile rear camera. can be taken using a reflector. In this way, the candidate can deploy the mobile camera and generate the most robust solution. No reflectors are needed when the upper part of the hand is withdrawn and the hand is placed under the mobile device. The location of the mobile device in this implementation allows the candidate to interact with the display of the mobile device for access control, point of sale (POS), and other applications requiring authentication.

Reference is now made to FIG. 15 , which presents a series of steps for authentication and registration. The candidate is instructed to capture several QR codes from a grid of 9 rectangles with different QR codes, based on the registration phase. A symbol display box on the device display 109 displays the view image. The symbol display boxes 1301, 1301a, 1301b, and 1301c display images captured in each of the four steps of authentication and registration. A timer symbol 1303 displays the length of capture time required for each step specified at the time of entry or the elapsed time since the start of the entire sequence. Display boxes of images captured at rectangles 1305, 1307, 1309, 1311 are organized with time stamps at 1303 for each step. In step 1, the candidate captures on display box 1301a only the RQ codes within rectangle 1305 and the left adjacent rectangle (part of 1307). In a second step performed within box display 1301b , the candidate captures the QR code within rectangles 1309 , 1311 as displayed at 1303 . In a third step, the candidate captures 4 QR codes from box 1307 on box 1301c for a length of 3 seconds. If all QR capture steps have been performed and aligned correctly in the right order within the required timing frame on the screen within box 1303, the candidate is authenticated.

Candidates can alternately capture and align grid rectangles of photos and symbols on cards, walls, signs and other usable surfaces. During enrollment, the user captures a sequence of grid pictures to each align with an on-screen grid or cell in any desired combination. Upon authentication, the candidate holds the mobile device and aligns the on-screen grid symbol with the grid picture as in the registration step sequence. The authentication phase may be synchronized with a timer or other display symbols measured on the screen.

The candidate may align the mobile camera with the viewing object, or the card may align with the camera. Additionally, any other combined fixed type mobile object, such as a card, may be part of the desired authentication sequence.

Another example is when there are 9 equal-sized squares arranged on the iPhone ^R ™ password screen. During verification, the candidate may present lines of one finger within a particular rectangle, then may present lines of each other within another particular rectangle, and so on. The combination of steps is stored in the registration step, and during verification the candidate must repeat the combination of steps. Therefore, the scammer cannot predict the combination of finger placement and selection of squares in the order performed during initial registration.

The on-screen symbols aligned with the viewing object may be predefined or manually defined in any order and arrangement. Symbols can be resized, reordered, moved, or handwritten by the user to allow the user to align the on-screen symbols with the viewing scene as was realized during listing. Candidates must select reproducible scenes for authentication to succeed.

For example, if a user selects a tree as part of a scene image, it is possible for the tree to change shape, rendering the image non-reproducible and ineligible for use.

Additionally, during registration, and upon subsequent authentication, the specified sequence of steps may combine any form of password-type keyboard typing. This password typing may be combined within the enrolment steps, either together or singly with the steps presented as described herein. For example, a candidate may enter a password in a specific timing sequence, at different timings of pressing each password letter, the length of time in the gap between each symbol, or some other combination. During authentication, a timer display, progress bar, or other equalizer shape (to show the desired measurement symbol press on each symbol, and any other combination of symbols for the keyboard symbol press time length and sequence timing of the gap between symbol presses) or otherwise a visual time display including a time display notification may be presented. In this way, even if the crook knows the password, he cannot use it because he does not know the exact sequence timing of the symbol presses. In this way, instead of using multiple passwords, the candidate could use the same sequence of symbols as the password, which would be difficult to crack due to the endless password-like permutations.

The keyboard sequence timing may be combined in time with all other sequence view image alignments and handwritten symbols or any combination with other presented authentication schemes.

It is believed that it will change the way people perform identification if users adopt an equally easy way to remember passwords while maintaining the highest level of security.

A crook who gets the password cannot use it unless he knows the exact sequence of steps (since many combinations are possible). Because the steps that follow are easily reproducible and intuitive, the candidate can create the complexity he demands, and the sequence may not be limited by a particular implementation, reproducible.

Reference is now made to FIG. 16 , which illustrates a handwritten candidate symbol, such as a signature biometric. During registration, in a first step, the candidate draws the registration symbol or signature shown at 1401 on the screen device 109 . Then, during the second stage, the candidate aligns any kind of auxiliary symbol or marker with the registered symbol or signature of the stage 1 1401 display symbol. Also during registration, in a second step, the candidate uses the display box 1403 to erase a portion of the signature. The auxiliary symbol may include a registered symbol or any symbol that aligns with the signature, and removes portions of the registered signature, any kind of distortion of the signature, and other symbols indicating portions of the signature.

The display during verification may have the secondary marker/hint of stage 2 of the listing 1403 display. The presented process enables high-level signature biometric authentication. The signature distortion of step 2 includes blurring, geometric distortion, or other image manipulation that stops the registration image reconstruction as in 1401 .

Another possible implementation includes the following steps:

In step 1 , the candidate places a signature or symbol on the display 109 in the symbol box 1401 . In step 2, the signature on the display box symbol 1401 is displayed on the screen 109, and the candidate inserts arbitrary symbols, scrambled lines, and other symbols to obscure the original signature in the symbol box display 1403 . Upon authentication, the candidate sees the scrambled line with the signature hidden within the display symbol 1403 . The candidate inserts the signature or symbol by following the lines of the hidden signature or symbol of entry step 1 ( 1401 ). In this implementation, the crook will have difficulty distinguishing the signature from the other hidden symbols of entry stage 2 1403 . This authentication method is difficult to solve within a short limited time frame. Handwritten marks or symbols may be combined with other forms of candidate authentication.

During authentication, a candidate is shown having one or more randomly selected enrollment images in a continuous manner in an unknown combination.

Additionally, a registered biometric method may include a set of registered image sequences or registered video sequences described herein. Upon authentication, the candidate must superimpose the viewing object with the on-screen listing video sequence or listing image sequence of the viewing object or symbol moving on the screen as presented within the listing.

For example, upon listing, the candidate may continuously overlap the selected travel line on the screen with the left edge of the driver's license. As an alternative, the candidate may have their driver's license superimposed with the moving line at a specific synchronization position of the line on the screen.

In another example, a candidate may define a set of symbols, such as lines or symbols, moving on a screen. During listing, two lines on the screen are rotating in different directions, and the candidate can superimpose the view object on one or more of the lines of movement at certain synchronization and the position of the lines on the screen. Upon authentication, two lines are displayed on the screen and the candidate will follow the registered synchronization position, time, relative to the object of view. Only the candidate will know the correct line alignment and synchronization between the line and the view object, and the candidate will be authenticated.

Candidates can alternately capture view images, aligning or applying symbols at specific locations and times relative to the view image as additional sequence steps in the authentication method.

In another example of a time display symbol, the display may have a watch-like display with a moving hand that the candidate may use with the time display synchronization method referred to herein. Additionally, the candidate may press or swipe the watch display when the moving hand of the watch reaches a predetermined position on the watch or captures a predetermined view image.

The display symbols may also consist of any scaled symbols of any kind, or any other quantity of measurement orientation symbols. For example, a candidate may use the display bar over a wide numerical range from one to one thousand. By sliding a finger on the screen bar, the candidate selects the desired value as an additional sequence step during authentication. The bars may be of any type, number of ranges, and precision required. Another symbol option could be any symbol with different parts of color and texture that the candidate can choose with an additional sequence step during registration and authentication.

Embodiments of the present invention may include general purpose or dedicated computer systems including various computer hardware components discussed in greater detail below. Embodiments within the scope of the present invention also include computer-readable media for carrying, or having, computer-executable instructions, computer-executable instructions, or data structures. Such computer-readable media can be any available media that can be accessed by a general purpose or special purpose computer system. By way of example, and not limitation, such computer readable media may be RAM, ROM, EPROM, flash disk, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage device, or to be accessed by a general purpose or dedicated computer system. may include non-transitory physical storage media, such as computer executable instructions, computer readable instructions, or other media that can be used to carry or store desired program code means in the form of data structures.

In the detailed description and in the claims that follow, a "computer system" is defined as one or more software modules, one or more hardware modules, or a combination thereof that work together to perform operations on electronic data. For example, the definition of a computer system includes the hardware components of a personal computer and software modules such as the operating system of the personal computer. The physical layout of the module is not important. A computer system may include one or more computers that are connected through a computer network. Likewise, a computer system may include a single physical device (such as a telephone or personal digital assistant “PDA”), in which case internal modules (such as memory and processor) work together to perform operations on electronic data. can Although a computer system may be mobile, the term "mobile computer system" includes, inter alia, laptop computers, notebook computers, cellular telephones, smart phones, cordless telephones, PDAs, portable computers with touch sensitive screens, and the like.

A "network" is defined as any architecture in which two or more computer systems can exchange data. The term "network" may include wide area networks, Internet local area networks, intranets, wireless networks such as "Wi-Fi™", virtual private networks, mobile access networks using the name of an access point, and the Internet. The data to be reformed may be in the form of an electrical signal that is meaningful to two or more computer systems. When data is transmitted or provided to a computer system or computer device over a network or other communication connection (eg, wired, wireless, or a combination of wired and wireless), such connection is classified as a transitory computer-readable medium. Accordingly, such connections are referred to as transitory computer-readable media. Combinations of the above should also be included within the scope of transitory computer-readable media and non-transitory computer-readable media. Computer-executable instructions include, for example, instructions and data that cause a general purpose computer system or special purpose computer system to perform a given function or group of functions.

As used herein, the term “server” refers to a computer system that includes a processor, data storage, and network adapters that are generally configured to provide services over a computer network. A computer system that receives the services provided by a server is known as a "client" computer system.

Claims (28)

A computerized biometric method comprising user registration and authentication, comprising:
The registration is:
capturing a pre-enrollment image of the view object while the user holds the view object in the field of view of the camera;
presenting the pre-registration image to a user on a display;
displaying the pre-registration image, causing a user to superimpose a symbol with an imaged feature of a view object displayed on the display;
superimposing the symbol with an imaged feature of an object of view on the display, storing the position and orientation of the symbol in symbol and image space and the pre-enrollment image as an enrollment image;
including,
The above certification is:
presenting on a display at least a portion of a previously stored registration image to a candidate, the previously stored registration image comprising the symbol;
instructing the candidate to present the viewing object to the camera while placing the viewing object in the field of view of the camera;
capturing time-series candidate images of the object of view with the camera while causing the candidate to align at least one candidate image of the object of view with a previously stored registration image or a portion of the registration image comprising a symbol; and
If at least one candidate image is aligned with the symbol in the registered image, authenticating the candidate as a previously registered user by verifying that the candidate image is an authentic image of the view object;
A computerized biometric recognition method comprising a. The computerized of claim 1 , wherein a viewing object located in the camera's field of view is fixed by a candidate, and wherein the alignment comprises distance in real space between the viewing object and the display and alignment of space and angle in image space. biometric identification method. The method of claim 1 , further comprising, during registration, selecting, by a user, the previously stored registration image from among a plurality of previously stored registration images. According to claim 1,
The registration is:
using the camera to capture registration images of the object of view and the background while the object of view is positioned in the field of view of the camera;
The above certification is:
capturing time-series candidate images of the candidate's view object and background; and
during said capture, displaying, on a display, time-series candidate images superimposed on enrollment images having respective backgrounds. The method of claim 1 , wherein the alignment is performed by instructing the candidate to align a feature of an object of view in a candidate image with a corresponding feature in the enrollment image. The method of claim 1 , wherein the verification is performed by comparing the enrollment image with the candidate image. The method of claim 1 , wherein the alignment comprises alignment in horizontal and vertical image coordinates, angular orientation in image space, and alignment in actual spatial distance to the camera. The computerized method of claim 1 , further comprising, prior to displaying at least a portion of the previously stored registered image, processing the registered image by performing an image distortion and manipulation procedure to deactivate reconstruction of the registered image. biometric identification method. According to claim 1,
instructing the candidate to align a feature of the object of view in the candidate image with a corresponding feature in the registration image; and
After the candidate aligns the feature of the object of view in the candidate image with the corresponding feature in the registered image, verifying that at least one of the candidate images is an authentic image of the object of view, so that the feature of the object of view corresponds to the corresponding feature in the registered image. and authenticating the candidate as a previously registered user by comparing with The computerized computerized method of claim 1 , further comprising, prior to overlapping, at least one of repositioning the symbol, hand-drawing the symbol, changing the size of the symbol, and rotating the symbol. biometric method. 3. The computerized method of claim 1, wherein presenting on the display to the candidate at least a portion of the previously stored registration image comprises presenting on the display at least a portion of an image of a stored symbol and the view object. biometric identification method. The method of claim 1, wherein the registration
presenting on a display a plurality of previously formed registration scales to the user;
selecting at least one of the scales displayed on the display, overlapping at least one of the registered images with the selected scale and changing the position, to generate the selected registration scale;
storing the selected scale as at least one registered image;
including,
The certification is
displaying, on the display, time-series candidate images superimposed on the selected registration scale;
aligning at least one of the candidate images with a registration scale selected by the candidate;
A computerized biometric recognition method comprising a. 13. The computer of claim 12, wherein during the authentication, presenting a registration scale to a candidate comprises displaying a plurality of registration scales, and wherein the method further comprises selecting a registration scale selected by the candidate. localized biometric methods. The registration image of claim 1 , wherein the previously stored registration image is retrieved from a previously electronically stored medium or the previously stored registration image is visually obtained from a photograph selected from photographs of a passport, identification card, driver's license and identification document using an imaging device. , a computerized biometric method. The method of claim 1 , wherein the prior enrollment image comprises at least partially handwritten password symbols. 16. The method of claim 15, wherein upon authentication, the candidate handwrites a handwritten password symbol written during registration. 17. The method of claim 16, wherein the enrollment image comprises handwritten symbols and previously formed symbols typed from a keyboard or entered using an input device. 2. The computerized method of claim 1, further comprising, during registration, applying a timed sequence of symbols by a user on a display, and further comprising, during authentication, instructing a candidate to repeat the timed sequence. biometric identification method. A mobile computer system for performing the computerized biometric identification method according to claim 1 , wherein a camera and a display are part of the mobile computer system. A non-transitory computer-readable medium encoded with processing instructions that cause a processor to execute the computerized biometric method of claim 1 . delete delete delete delete delete delete delete delete

Images