IP Library › Granted Patent US 12,256,020
Granted Patent B1
US 12,256,020 · App. 18/750,590 · Granted Mar 18, 2025

Systems and methods for generating attested data

Inventors: Yin Yu Michelle Lai (Singapore, SG); Alan Fairless (Lenexa, KS); Sean Mollet (Overland Park, KS)
Assignee: Purple Squid LLC
H04L9/3247G06V10/44G06V10/764H04L9/0825H04L9/3221
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 12,256,020
App. No.
18/750,590
Granted
Mar 18, 2025
Kind
B1
Abstract

In various embodiments, systems and methods for generating attested video image data are provided. In some embodiments, an ISP pipeline incorporates a secure cryptographic signing mechanism to digitally sign content-based outputs in a way that attests to the authenticity of the process by which the content was processed before output from the camera. The ISP pipeline may include a hardware security module that inputs video image data from an image sensor, and generates an attestation package that includes attestations about the video content, in combination with an attestation of how it was produced, and signs those attestations together using a key to produce an output comprising attested media data. The ISP pipeline may apply supplemental attestations to the attestation package representing facts associated with the video content that evidence the authenticity. The digital signature applied to the attestation package may use a zero-knowledge proof to describe the signing key.

Claims (42)

1. A device comprising:

at least one image sensor associated with a hardware device identifier;

a hardware security module coupled to the at least one image sensor by at least one data bus, the hardware security module comprising a secure enclave that includes:

a first controller;

a memory management unit coupled to a secure memory within the secure enclave of the hardware security module, wherein the first controller communicates with the secure memory using the memory management unit; and

a digital signature engine;

wherein the hardware security module, using the memory management unit, stores to the secure memory a trusted image data bundle received from the at least one image sensor via the at least one data bus, wherein the trusted image data bundle includes image data captured by the at least one image sensor and the hardware device identifier;

wherein the first controller is configured to generate within the secure memory an attestation package comprising at least a representation of the image data and a representation of the hardware device identifier;

wherein the digital signature engine is configured to apply a cryptographic signature to the attestation package using a signature key, wherein the signature key is associated with a video session comprising a video image sequence of the image data and when the video session is started, a zero-knowledge proof is created showing that the signature key is signed by a key in a published database of authentic hardware keys, without revealing identifying information associated with the device; and

wherein the hardware security module is configured to output a set of attested media data based on the attestation package.

2. The device of claim 1 , wherein the first controller adds one or more supplemental attestation elements to the attestation package, wherein the one or more supplemental attestation elements represent one or more facts associated with the image data.

3. The device of claim 2 , wherein the one or more supplemental attestation elements comprise one or more of: time data, location data, satellite navigation system data, a pseudo-random-noise (PRN) code, an asymmetric key, and data received from a blockchain or blockchain related technology.

4. The device of claim 2 , wherein the first controller generates an image data pipeline description that identifies one or more processes applied to the image data to produce the representation of the image data, wherein the attestation package further comprises the image data pipeline description.

5. The device of claim 1 , wherein the hardware security module is further coupled to a second controller external to the secure enclave, wherein the first controller adds one or more supplemental attestation elements to the attestation package based on an indication from the second controller.

6. The device of claim 1 , wherein the hardware security module further comprises a video encoder to encode the image data based on a codec to produce an encoded video data, wherein the representation of the image data comprises the encoded video data.

7. The device of claim 1 , wherein the hardware security module further comprises one or more machine learning models that generate a tokenization of one or more features extracted from the image data, wherein the first controller generates the representation of the image data as a computer vision rendering of the image data based at least on the tokenization of the one or more features.

8. The device of claim 1 , wherein the hardware security module further comprises one or more machine learning models that generate at least a classification characterizing one or more features extracted from the image data, wherein the representation of the image data comprises the classification characterizing one or more features extracted from the image data.

9. The device of claim 1 , wherein the at least one image sensor authenticates with the hardware security module based at least on an authenticated key exchange.

10. The device of claim 1 , wherein the at least one data bus comprises at least one of an I 2 C protocol channel and a MIPI protocol channel.

11. The device of claim 1 , wherein the hardware security module is configured to generate metadata comprising a precommitment of the attestation package for submission to a blockchain or blockchain related technology.

12. A device comprising:

at least one image sensor associated with a hardware device identifier; and

a processing device coupled to the at least one image sensor, wherein the processing device comprises a secure enclave that includes an attestation package assembly function, wherein the attestation package assembly function receives a trusted image data bundle from the at least one image sensor, wherein the trusted image data bundle includes image data captured by the at least one image sensor and the hardware device identifier;

wherein the attestation package assembly function is configured to generate an attestation package comprising a plurality of attestations that include at least a representation of the image data and a representation of the hardware device identifier, and to apply a cryptographic signature to the attestation package using a signature key, wherein the signature key is associated with a video session comprising a video image sequence of the image data and when the video session is started, a zero-knowledge proof is created showing that the signature key is signed by a key in a published database of authentic hardware keys, without revealing identifying information associated with the device; and

wherein the processing device comprises an interface configured to output a set of attested media data based on the attestation package.

13. The device of claim 12 , wherein the attestation package comprises a tree structure comprising data fields representing individual attestations of the plurality of attestations.

14. The device of claim 12 , wherein the attestation package further comprises one or more supplemental attestation elements, each representing information associated with one or more of: the image data, the capturing of the image data, the detection of features from the image data, the classification of features from the image data, and one or more processes performed to produce the representation of the image data.

15. A system comprising:

at least one memory component; and

one or more processing devices coupled to the at least one memory component, the one or more processing devices to perform operations comprising:

receiving a trusted image data bundle from an image sensor, wherein the trusted image data bundle includes image data captured by at least one image sensor and a hardware device identifier associated with the image sensor, wherein the hardware device identifier includes at least a first public key associated with the image sensor, and a second public key associated with the one or more processors;

generating an attestation package comprising a plurality of attestations that includes at least a representation of the image data and a representation of the hardware device identifier; and

generating an output comprising a set of attested media data based on applying a digital signature to the attestation package using a signature key, wherein the signature key is associated with a video session comprising a video image sequence of the image data and when the video session is started, a zero-knowledge proof is created showing that the signature key is signed by a key in a published database of authentic hardware keys, without revealing identifying information associated with the system.

16. The system of claim 15 , the operations further comprising:

based on the image data, generating the representation of the image data to comprise at least one of:

encoded video data based on the image data;

a computer vision rendering of the image data based at least on one or more features extracted from the image data by one or more machine learning models; and

one or more classifications characterizing the one or more features extracted from the image data by the one or more machine learning models; and

wherein when the representation of the image data does not include the encoded video data, the plurality of attestations comprises at least one attestation confirming that the representation of the image data does not include the encoded video data.

17. The system of claim 15 , the operations further comprising:

applying a selective disclosure redaction of one or more of the plurality of attestations to generate a distribution version of the set of attested media data for presentation by a presentation application; and

wherein the presentation application validates the distribution version of the set of attested media data using one or more validation resources based at least on the signature key.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 30, 2024
From: FAIRLESS, ALAN; MOLLET, SEAN; LAI, YIN YU MICHELLE
To: PURPLE SQUID LLC
Reel/Frame 068449/0527 →
References Cited (31)
US 9819941B2 · Ramasubramonian · 2017 [cited by examiner]
US 11449584B1 · Pamucci · 2022 [cited by applicant]
US 11770260B1 · Pamucci · 2023 [cited by applicant]
US 12003622B1 · Anderson · 2024 [cited by examiner]
US 20040022523A1 · Duerr et al. · 2004 [cited by applicant]
US 20080260161A1 · Yokota · 2008 [cited by examiner]
US 20120198239A1 · Oishi · 2012 [cited by applicant]
US 20170373859A1 · Shors et al. · 2017 [cited by applicant]
US 20180253567A1 · Gonzalez-Banos et al. · 2018 [cited by applicant]
US 20190349204A1 · Enke · 2019 [cited by examiner]
US 20200302133A1 · Ciet · 2020 [cited by examiner]
US 20220067129A1 · Kumar · 2022 [cited by applicant]
US 20220375298A1 · Schertzer · 2022 [cited by examiner]
US 20230385385A1 · Ur · 2023 [cited by applicant]
US 20240007292A1 · Moon · 2024 [cited by examiner]
US 20240056310A1 · Mahara · 2024 [cited by examiner]
US 20240205012A1 · Jordan · 2024 [cited by examiner]
US 20240313961A1 · Li · 2024 [cited by examiner]
CN 116318880A · 2023 [cited by applicant]
EP 3537319A1 · 2019 [cited by applicant]
JP 09200730A · 1997 [cited by applicant]
JP 2000175055A · 2000 [cited by applicant]
JP 2001292360A · 2001 [cited by applicant]
WO 2010001282A2 · 2010 [cited by applicant]
WO 2020037400A1 · 2020 [cited by applicant]
Amd Xilinx : “Vivado Design Suite User Guide : Programming and Debugging (UG908)”, Bitstream Verify and Readback for FPGAs and MPSoCs, May 30, 2024, 5 pages. [cited by applicant]
Boneh, D., “The Zero Knowledge Podcast—Episode 256 : New ZK Use Cases with Dan Boneh”, Available at : <https://youtu.be/EKoY8ysGblk?si=tpgA8sDheSgmMlvm&t=920>, Nov. 30, 2022, 4 pages. [cited by applicant]
Datta et al., “Using ZK Proofs to Fight Disinformation”, Medium, Sep. 30, 2022. 12 pages. [cited by applicant]
Ikonomou, T., “Defence force top brass shares fears about AI causing ‘truth decay’”, INQueensland, Sep. 15, 2023, 8 pages. [cited by applicant]
Kang et al., “Fighting AI-generated Audio with Attested Microphones and ZK-SNARKs: the Attested Audio Experiment”, Medium, Jun. 13, 2023, pp. 1-9. [cited by applicant]
Vigliarolo, B., “Sony camera feature hopes to make digital images immune to secret manipulation”, The Register—Personal Tech, Aug. 15, 2022. 3 pages. [cited by applicant]
Cited By (1)
US 12,657,344