IP Library Granted Patent US 12,316,624
Granted Patent B2
US 12,316,624 · App. 18/097,738 · Granted May 27, 2025

Screen-analysis based device security

Inventors: Matthew Sharifi (Kilchberg, CH); Kai Wang (Santa Monica, CA); David Petrou (Brooklyn, NY)
Assignee: GOOGLE LLC
H04L63/083G06F21/316G06F21/36G06N20/00H04W4/029H04W12/06H04W12/126H04W12/30H04L63/102H04W4/80H04W12/10H04W12/63H04W12/68
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,316,624
App. No.
18/097,738
Granted
May 27, 2025
Kind
B2
Abstract

Systems and methods are provided for a content-based security for computing devices. An example method includes identifying content rendered by a mobile application, the content being rendered during a session, generating feature vectors from the content and determining that the feature vectors do not match a classification model. The method also includes providing, in response to the determination that the feature vectors do not match the classification model, a challenge configured to authenticate a user of the mobile device. Another example method includes determining a computing device is located at a trusted location, capturing information from a session, the information coming from content rendered by a mobile application during the session, generating feature vectors for the session, and repeating this until a training criteria is met. The method also includes training a classification model using the feature vectors and authenticating a user of the device using the trained classification model.

Claims (73)

1. A method implemented by one or more processors, the method comprising:

identifying a user profile that corresponds to a user and that is active for an electronic device;

determining whether current activity of the electronic device, while the user profile is active for the electronic device, matches the user profile, wherein determining whether the current activity of the electronic device matches the user profile comprises:

processing, using a classification model, feature vectors that are generated based on content that is displayed on a screen of the electronic device during the current activity of the electronic device to determine whether the feature vectors corresponding to the content that is displayed on the screen of the electronic device during the current activity of the electronic device match the user profile; and

in response to determining the feature vectors corresponding to the current activity of the electronic device do not match the user profile:

determining whether the current activity of the electronic device, while the user profile is active for the electronic device, matches an additional user profile, of a plurality of additional user profiles, the additional user profile corresponding to an additional user; and

in response to determining the current activity of the electronic device matches the additional user profile:

initiating a profile switch, to the additional user profile, at the electronic device.

2. The method of claim 1 , wherein determining whether the current activity of the electronic device matches the additional user profile comprises:

processing, using the classification model, the feature vectors corresponding to the current activity of the electronic device to determine whether the feature vectors corresponding to the current activity of the electronic device match the additional user profile.

3. The method of claim 1 , further comprising:

in response to determining the feature vectors corresponding to the current activity of the electronic device do not match the additional user profile:

determining whether the current activity of the electronic device, while the user profile is active for the electronic device, matches a further additional user profile, of the plurality of additional user profiles, the further additional user profile corresponding to a further additional user; and

in response to determining the current activity of the electronic device matches the further additional user profile:

initiating the profile switch, to the further additional user profile, at the electronic device.

4. The method of claim 3 , wherein determining whether the current activity of the electronic device matches the further additional user profile comprises:

processing, using the classification model, the feature vectors corresponding to the current activity of the electronic device to determine whether the feature vectors corresponding to the current activity of the electronic device match the further additional user profile.

5. The method of claim 1 , wherein determining whether the current activity of the electronic device matches the user profile is in response to determining that the current activity at the electronic device corresponds to outlier activity for the user profile.

6. The method of claim 1 , wherein initiating the profile switch at the electronic device comprises:

automatically signing out of the user profile to deactivate the user profile at the electronic device; and

automatically signing in to the additional user profile to activate the additional user profile at the electronic device.

7. The method of claim 1 , wherein initiating the profile switch at the electronic device comprises:

automatically signing out of the user profile to deactivate the user profile at the electronic device; and

causing a sign-in user interface to be visually rendered at the electronic device to enable the additional user to sign-in to the additional user profile.

8. The method of claim 7 , further comprising:

in response to visually rendering the sign-in user interface at the electronic device:

receiving, at the electronic device, credentials to sign-in to the additional user profile; and

signing in to the additional user profile based on the credentials.

9. The method of claim 1 , wherein initiating the profile switch at the electronic device comprises:

causing a prompt to be visually rendered at the electronic device to request permission to switch to the additional user profile associated with the additional user.

10. The method of claim 9 , further comprising:

in response to visually rendering the prompt at the electronic device:

receiving, at the electronic device, user input to permit switching to the additional user profile.

11. The method of claim 1 , wherein generating the feature vectors based on the content that is displayed on the screen of the electronic device comprises:

processing at least a portion of the content that is displayed on the screen of the electronic device;

identifying one or more features included in the content that is displayed on the screen of the electronic device; and

generating the feature vectors using one or more of the features included in the content that is displayed on the screen of the electronic device.

12. A system comprising:

at least one processor; and

at least one memory storing instructions that, when executed, cause the at least one processor to:

identify a user profile that corresponds to a user and that is active for an electronic device;

determine whether current activity of the electronic device, while the user profile is active for the electronic device, matches the user profile, wherein the instructions to determine whether the current activity of the electronic device matches the user associated with the user profile comprise instructions to:

process, using a classification model, feature vectors that are generated based on content that is displayed on a screen of the electronic device during the current activity of the electronic device to determine whether the feature vectors corresponding to the content that is displayed on the screen of the electronic device during the current activity of the electronic device match the user profile;

in response to determining the feature vectors corresponding to the current activity of the electronic device do not match the user profile:

determine whether the current activity of the electronic device, while the user profile is active for the electronic device, matches an additional user profile, of a plurality of additional user profiles, the additional user profile corresponding to an additional user; and

in response to determining the current activity of the electronic device matches the additional user profile:

initiate a profile switch, to the additional user profile, at the electronic device.

13. The system of claim 12 , wherein the instructions to determine whether the current activity of the electronic device matches the additional user profile comprise instructions to:

process, using the classification model, the feature vectors corresponding to the current activity of the electronic device to determine whether the feature vectors corresponding to the current activity of the electronic device match the additional user profile.

14. The system of claim 12 , wherein the instructions further comprise instructions to:

in response to determining the feature vectors corresponding to the current activity of the electronic device do not match the additional user profile:

determine whether the current activity of the electronic device, while the user profile is active for the electronic device, matches a further additional user profile, of the plurality of additional user profiles, the further additional user profile corresponding to a further additional user; and

in response to determining the current activity of the electronic device matches the further additional user profile:

initiate the profile switch, to the further additional user profile, at the electronic device.

15. The system of claim 14 , wherein the instructions to determine whether the current activity of the electronic device matches the further additional user profile comprise instructions to:

process, using the classification model, the feature vectors corresponding to the current activity of the electronic device to determine whether the feature vectors corresponding to the current activity of the electronic device match the further additional user profile.

16. The system of claim 12 , wherein determining whether the current activity of the electronic device matches the user profile is in response to determining that the current activity at the electronic device corresponds to outlier activity for the user profile.

17. The system of claim 12 , wherein the instructions to initiate the profile switch at the electronic device comprise instructions to:

automatically sign-in of the user profile to deactivate the user profile at the electronic device; and

automatically sign-in to the additional user profile to activate the additional user profile at the electronic device.

18. The system of claim 12 , wherein the instructions to initiate the profile switch at the electronic device comprise instructions to:

automatically sign out of the user profile to deactivate the user profile at the electronic device; and

cause a sign-in user interface to be visually rendered at the electronic device to enable the additional user to sign-in to the additional user profile.

19. The system of claim 12 , wherein the instructions to initiate the profile switch at the electronic device comprise instructions to:

cause a prompt to be visually rendered at the electronic device to request permission to switch to the additional user profile associated with the additional user.

20. A non-transitory computer-readable storage medium storing instructions that, when executed, cause at least one processor to execute the instructions, the instructions causing the at least one processor to:

identify a user profile that corresponds to a user and that is active for an electronic device;

determine whether current activity of the electronic device, while the user profile is active for the electronic device, matches the user profile, wherein the instructions to determine whether the current activity of the electronic device matches the user profile comprise instructions to:

process, using a classification model, feature vectors that are generated based on content that is displayed on a screen of the electronic device during the current activity of the electronic device to determine whether the feature vectors corresponding to the content that is displayed on the screen of the electronic device during the current activity of the electronic device match the user profile;

in response to determining the feature vectors corresponding to the current activity of the electronic device do not match the user profile:

determine whether the current activity of the electronic device, while the user profile is active for the electronic device, matches an additional user profile, of a plurality of additional user profiles, the additional user profile corresponding to an additional user; and

in response to determining the current activity of the electronic device matches the additional user profile:

initiating a profile switch, to the additional user profile, at the electronic device.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 14, 2023
From: SHARIFI, MATTHEW; WANG, KAI; PETROU, DAVID
To: GOOGLE INC.
Reel/Frame 062693/0383 →
CHANGE OF NAME Recorded Feb 14, 2023
From: GOOGLE INC.
To: GOOGLE LLC
Reel/Frame 062749/0030 →
Continuity (4)
Continuation 17064213 · Oct 6, 2020
Continuation 16133395 · Sep 17, 2018
Continuation 14739107 · Jun 15, 2015
Related Publication 20230156000A1 · May 18, 2023
References Cited (88)
US 7162522B2 · Adar et al. · 2007 [cited by applicant]
US 8352386B2 · Chitiveli et al. · 2013 [cited by applicant]
US 8468110B1 · Podgorny et al. · 2013 [cited by applicant]
US 8571319B2 · Balasubramanian et al. · 2013 [cited by applicant]
US 8811727B2 · Mohamed · 2014 [cited by applicant]
US 8928587B1 · Smus · 2015 [cited by applicant]
US 8978111B2 · Doris-Down et al. · 2015 [cited by applicant]
US 8984591B2 · McFarland · 2015 [cited by applicant]
US 9165406B1 · Gray et al. · 2015 [cited by applicant]
US 9185095B1 · Moritz et al. · 2015 [cited by applicant]
US 9405377B2 · Li · 2016 [cited by applicant]
US 9582482B1 · Sharifi et al. · 2017 [cited by applicant]
US 9590966B2 · Sheller et al. · 2017 [cited by applicant]
US 9959506B1 · Karppanen · 2018 [cited by examiner]
US 10078803B2 · Sharifi et al. · 2018 [cited by applicant]
US 10489462B1 · Rogynskyy · 2019 [cited by applicant]
US 10803408B2 · Sharifi et al. · 2020 [cited by applicant]
US 20060179410A1 · Deeds · 2006 [cited by applicant]
US 20070168379A1 · Patel et al. · 2007 [cited by applicant]
US 20080083040A1 · Dani et al. · 2008 [cited by applicant]
US 20080301101A1 · Baratto et al. · 2008 [cited by applicant]
US 20090031227A1 · Chakrabarti · 2009 [cited by applicant]
US 20090063431A1 · Erol et al. · 2009 [cited by applicant]
US 20090170552A1 · Lin · 2009 [cited by applicant]
US 20090216563A1 · Sandoval · 2009 [cited by applicant]
US 20090253455A1 · Huang · 2009 [cited by applicant]
US 20090328175A1 · Shuster · 2009 [cited by applicant]
US 20100169176A1 · Turakhia · 2010 [cited by applicant]
US 20110128288A1 · Petrou et al. · 2011 [cited by applicant]
US 20110131235A1 · Petrou et al. · 2011 [cited by applicant]
US 20120060214A1 · Nahari · 2012 [cited by applicant]
US 20120083294A1 · Bray et al. · 2012 [cited by applicant]
US 20130042327A1 · Chow · 2013 [cited by applicant]
US 20130067261A1 · Carroll · 2013 [cited by applicant]
US 20130088650A1 · Rouady · 2013 [cited by examiner]
US 20130091463A1 · Nordstrom et al. · 2013 [cited by applicant]
US 20130108161A1 · Carr · 2013 [cited by applicant]
US 20130191908A1 · Klein · 2013 [cited by applicant]
US 20130311288A1 · Taylor · 2013 [cited by applicant]
US 20140066020A1 · Sadhvani et al. · 2014 [cited by applicant]
US 20140109018A1 · Casey et al. · 2014 [cited by applicant]
US 20140280587A1 · Sharma · 2014 [cited by applicant]
US 20140283014A1 · Tse et al. · 2014 [cited by applicant]
US 20140333524A1 · Liu et al. · 2014 [cited by applicant]
US 20140337243A1 · Dutt et al. · 2014 [cited by applicant]
US 20140372430A1 · Sitruk · 2014 [cited by examiner]
US 20150067845A1 · Chari · 2015 [cited by examiner]
US 20150355955A1 · Chakra et al. · 2015 [cited by applicant]
US 20160055246A1 · Marcin et al. · 2016 [cited by applicant]
US 20160182502A1 · Smith et al. · 2016 [cited by applicant]
US 20160321052A1 · Sharifi · 2016 [cited by examiner]
US 20200372016A1 · Rogynskyy · 2020 [cited by applicant]
US 20210019660A1 · Sharifi et al. · 2021 [cited by applicant]
US 20220147602A1 · Streit · 2022 [cited by applicant]
CN 1714356 · 2005 [cited by applicant]
CN 1943240 · 2007 [cited by applicant]
CN 101405743 · 2009 [cited by applicant]
CN 104268481 · 2015 [cited by applicant]
CN 103533546 · 2017 [cited by applicant]
CN 103530543 · 2017 [cited by applicant]
CN 104765995 · 2018 [cited by applicant]
EP 2579588 · 2013 [cited by applicant]
JP 2008084125 · 2008 [cited by applicant]
JP 2012203820 · 2012 [cited by applicant]
WO 2012075315 · 2012 [cited by applicant]
European Patent Office; Intention to Grant issued in Application No. 20199911.7, 44 pages. dated Aug. 17, 2022. [cited by applicant]
German Patent Office; Office Action issued in Application No. 112016002708.7, 16 pages, dated Feb. 3, 2022. [cited by applicant]
Intellectual Property Office of Great Britain; Examination Report issued in Application No. 1714990.7; 4 pages; dated May 17, 2021. [cited by applicant]
European Patent Office; Communication issued in Application No. 20199911.7; 10 pages; dated Feb. 22, 2021. [cited by applicant]
China National Intellectual Property Administration; Notice of Allowance issue in Application No. 201680019373.0, 4 pages, dated Oct. 29, 2020. [cited by applicant]
Chinese Patent Office; Office Action issued in Application No. 201680019373.0 dated Jul. 14, 2020. [cited by applicant]
European Patent Office; Intention to Grant issued in Application No. 16734091.8; 50 pages; dated May 18, 2020. [cited by applicant]
China National Intellectual Property Administration; Office Action, Application No. 201680019373.0, 43 pages, dated Nov. 4, 2019. [cited by applicant]
Bassu; A New Mobile Biometric Based Upon Usage Context, 2013 IEEE International Conference on Technologies for Homeland Security (HST); pp. 441-446; dated Nov. 12, 2013. [cited by applicant]
Roy, et al., An HMM-based Behavior Modeling Approach for Continuous Mobile Authentication, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP); pp. 3789-3793; dated May 4, 2014. [cited by applicant]
Shi, et al., Implicit Authentication through Learning User Behavior, Information Security, Springer Berlin Heidelberg; pp. 99-113; dated Oct. 25, 2010. [cited by applicant]
Korean Intellectual Property Office; Allowance of Patent issued in Application No. 10-2017-7027716 dated Aug. 27, 2018. [cited by applicant]
Korean Intellectual Property Office; Office Action issued in Appl. No. 10-2017-7027716 dated Feb. 14, 2018. [cited by applicant]
Japanese Intellectual Property Office; Allowance of Patent issued in Application No. 2017550858 dated Oct. 22, 2018. [cited by applicant]
International Search Report and Written Opinion from PCT/US2016037210; 14 pages; dated Aug. 23, 2016. [cited by applicant]
European Patent Office; Examination Report issued in Application No. 16734091.8 dated May 15, 2019. [cited by applicant]
China National Intellectual Property Administration; Notification of First Office Action issued in Application No. 202011567283.X; 16 pages; dated Dec. 29, 2023. [cited by applicant]
Roy, Aditi et al.; An HMM-Based Behavior Modeling Approach for Continuou S Mobile Authentic Ation; 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP); 5 pages; dated Jul. 14, 2014. [cited by applicant]
Shi, E. et al.; Implicit Authentication through Learning User Behavior; ICS 2010: Information Security: 16 pages; dated Oct. 25, 2011. [cited by applicant]
Intellectual Property India; Hearing Notice issued in Application No. 201747034118; 2 pages; dated Jan. 9, 2024. [cited by applicant]
Min; Access Identity Authentication Method of Desktop Cloud; 5 pages; dated Jul. 10, 2013. [cited by applicant]
Zhou, E., et al.; Screen Image Capture Technology Based on Mirror Driver; 6 pages; dated Jun. 15, 2013. [cited by applicant]
China National Intellectual Property Administration; Notice of Grant issued in Application No. 202011567283.X; 7 pages; dated May 20, 2024. [cited by applicant]