Granted Patent B2

US 12,585,747 · App. 18/797,282 · Granted Mar 24, 2026

Inventor: Justin Douglas Eltoft (Pleasant Prairie, WI)

Assignee: Google LLC

G06F21/32G06F3/04883G06V40/13G06V40/40

Abstract

Claims (60)

1 . A computer-implemented method comprising:

obtaining a fingerprint image from a user using a fingerprint sensor of a device;

analyzing the fingerprint image for fingerprint characteristics and determining one or more fingerprint detection scores based on the analyzing, wherein each fingerprint detection score is within a respective score range that includes a respective fingerprint validity threshold;

in response to the one or more fingerprint detection scores meeting the respective fingerprint validity threshold by greater than a respective particular range of the respective fingerprint validity threshold, providing a positive authentication indication;

obtaining respective sensor data from each of a plurality of additional sensors of the device, wherein the plurality of additional sensors include at least one additional sensor selected from the group of: an inertial measurement unit (IMU), an accelerometer, a gyroscope, an ambient light sensor, a proximity sensor, face detection sensor, a touchscreen, and combinations thereof; and

in response to at least one of the one or more fingerprint detection scores meeting the respective fingerprint validity threshold within the respective particular range of the respective fingerprint validity threshold:

analyzing the respective sensor data of each additional sensor to determine a respective context score for the additional sensor;

determining whether each of the respective context scores meets a respective context validity threshold associated with a type of the respective sensor data;

in response to determining that the respective context scores meet the respective context validity thresholds, providing the positive authentication indication; and

in response to determining that at least one context score does not meet the respective context validity threshold, providing a negative authentication indication.

2 . The computer-implemented method of claim 1 , wherein obtaining the respective sensor data from each of the plurality of additional sensors includes performing the obtaining the respective sensor data in response to the at least one of the one or more fingerprint detection scores meeting the respective fingerprint validity threshold within the respective particular range of the fingerprint validity threshold.

3 . The computer-implemented method of claim 1 wherein analyzing the fingerprint image for fingerprint characteristics includes analyzing the fingerprint image for spoof characteristics using a spoof detection technique, and wherein the one or more fingerprint detection scores includes a spoof score from the spoof detection technique.

4 . The computer-implemented method of claim 1 , wherein analyzing the fingerprint image for fingerprint characteristics includes comparing the fingerprint image to one or more profile images of the user using a matching fingerprint technique, wherein the one or more fingerprint detection scores includes a match score from the matching fingerprint technique.

5 . The computer-implemented method of claim 1 , wherein at least one of the context scores is based on one or more patterns in the respective sensor data detected within a particular time period prior to a time of obtaining the fingerprint image.

6 . The computer-implemented method of claim 1 , wherein one of the additional sensors is the IMU, the accelerometer, or the gyroscope, and wherein one of the context scores is based on motion of the device in a motion pattern indicating that the motion is caused by the user.

7 . The computer-implemented method of claim 6 , wherein the motion pattern indicating that the motion is caused by the user is based on historical sensor data indicating past motions of the device as detected by the IMU, the accelerometer, or the gyroscope.

8 . The computer-implemented method of claim 6 , wherein one of the additional sensors is the ambient light sensor, and wherein a second one of the context scores is based on illumination detected by the ambient light sensor in an illumination pattern indicating use of the device by the user.

9 . The computer-implemented method of claim 8 , wherein the illumination pattern indicating use of the device by the user is based on historical sensor data indicating past illumination detected by the ambient light sensor during previous use of the device.

10 . The computer-implemented method of claim 1 , wherein one of the additional sensors is the touchscreen, and wherein one of the context scores is based on one or more touch gestures detected on the touchscreen within a particular time period prior to a time of obtaining the fingerprint image.

11 . The computer-implemented method of claim 1 , wherein one of the additional sensors is the proximity sensor, and wherein one of the context scores is based on whether a proximity detection was detected within a particular time period prior to a time of obtaining the fingerprint image.

12 . The computer-implemented method of claim 1 , further comprising detecting sensor data from another additional sensor of the device if the respective context scores are within a low confidence range of the respective context validity thresholds.

13 . The computer-implemented method of claim 1 , wherein analyzing the respective sensor data of each additional sensor and determining whether each of the respective context scores meets the respective context validity threshold are performed in response to:

the at least one fingerprint detection score being within the respective particular range of the respective fingerprint validity threshold, and

at least one failed user attempt to be authenticated by the device being detected within a particular time period prior to a time of obtaining the fingerprint image.

14 . The computer-implemented method of claim 1 , wherein analyzing the respective sensor data of each additional sensor and determining whether each of the respective context scores meets the respective context validity threshold are performed in response to:

the at least one fingerprint detection score being within the respective particular range of the respective fingerprint validity threshold, and

the at least one fingerprint detection score being lower than one or more previous fingerprint detection scores determined for previous fingerprint images captured by the fingerprint sensor.

15 . The computer-implemented method of claim 1 , wherein the context validity thresholds are adjustable and are based on at least one condition selected from the group of:

a number of user attempts to be authenticated by the device within a particular time period prior to a time of obtaining the fingerprint image;

a number of unlocks of the device within a particular time period prior to a time of obtaining the fingerprint image; and

combinations thereof.

16 . A mobile device comprising:

at least one processor;

a fingerprint sensor coupled to the at least one processor, wherein the fingerprint sensor is configured to provide sensor data to the at least one processor; and

a plurality of additional sensors coupled to the at least one processor, wherein the one or more additional sensors are configured to provide sensor data to the at least one processor, wherein the additional sensors include at least one additional sensor selected from the group of: an inertial measurement unit (IMU), an accelerometer, a gyroscope, an ambient light sensor, a proximity sensor, face detection sensor, a touchscreen, and combinations thereof;

wherein the at least one processor is configured to perform operations including:

obtaining a fingerprint image from a user using the fingerprint sensor;

analyzing the fingerprint image for fingerprint characteristics and determining one or more fingerprint detection scores based on the analyzing, wherein each fingerprint detection score is within a respective score range that includes a respective fingerprint validity threshold;

in response to the one or more fingerprint detection scores meeting the respective fingerprint validity threshold by greater than a respective particular range of the respective fingerprint validity threshold, providing a positive authentication indication;

obtaining respective sensor data from each of the additional sensors; and

in response to at least one of the one or more fingerprint detection scores meeting the respective fingerprint validity threshold within the respective particular range of the respective fingerprint validity threshold:

analyzing the respective sensor data of each additional sensor to determine a respective context score for the additional sensor;

determining whether each of the respective context scores meets a respective context validity threshold associated with a type of the respective sensor data;

in response to determining that the respective context scores meet the respective context validity thresholds, accepting the fingerprint image as authentic; and

in response to determining that at least one context score does not meet the respective context validity threshold, providing a negative authentication indication.

17 . The mobile device of claim 16 , wherein at least one of the context scores is based on one or more patterns in the sensor data detected within a particular time period prior to a time of obtaining the fingerprint image.

18 . The mobile device of claim 16 , wherein one of the additional sensors is the IMU, the accelerometer, or the gyroscope, and wherein one of the context scores is based on motion of the device in a motion pattern indicating that the motion is caused by the user.

19 . The mobile device of claim 16 , wherein the operations of analyzing the sensor data of each additional sensor and determining whether each of the respective context scores meets the respective context validity threshold are performed in response to:

the at least one fingerprint detection score being within the respective particular range of the respective fingerprint validity threshold, and

at least one failed user attempt to be authenticated by the device being detected within a particular time period prior to a time of obtaining the fingerprint image.

20 . A non-transitory computer-readable medium with instructions stored thereon that, when executed by a processor, cause the processor to perform operations comprising:

obtaining a fingerprint image from a user using a fingerprint sensor of a device;

analyzing the fingerprint image for fingerprint characteristics and determining one or more fingerprint detection scores based on the analyzing, wherein each fingerprint detection score is within a respective score range that includes a respective fingerprint validity threshold;

in response to the one or more fingerprint detection scores meeting the respective fingerprint validity threshold by greater than a respective particular range of the respective fingerprint validity threshold, providing a positive authentication indication;

obtaining respective sensor data from each of a plurality of additional sensors of the device, wherein the plurality of additional sensors include at least one additional sensor selected from the group of: an inertial measurement unit (IMU), an accelerometer, a gyroscope, an ambient light sensor, a proximity sensor, face detection sensor, a touchscreen, and combinations thereof; and

in response to at least one of the one or more fingerprint detection scores meeting the respective fingerprint validity threshold within the respective particular range of the respective fingerprint validity threshold:

analyzing the respective sensor data of each additional sensor to determine a respective context score for the additional sensor;

determining whether each of the respective context scores meets a respective context validity threshold associated with a type of the respective sensor data;

in response to determining that the respective context scores meet the respective context validity thresholds, accepting the fingerprint image as authentic; and

in response to determining that at least one context score does not meet the respective context validity threshold, providing a negative authentication indication.

Assignments (1)

Continuity (1)

References Cited (10)