US-12619749-B2 - System for image/video authenticity verification

US12619749B2US 12619749 B2US12619749 B2US 12619749B2US-12619749-B2

Abstract

Embodiments of the invention are directed to systems, methods, and computer program products for intelligent verification of digital files via the analysis of metadata and other file characteristics. The system is adaptive, in that it can be adjusted based on the needs or goals of the user utilizing it, or may intelligently and proactively adapt based on the files or data received for processing. The system may be seamlessly embedded within existing applications or programs that the user may already use to interact with one or more entities.

Inventors

Joseph Benjamin Castinado
Naoll Addisu Merdassa

Assignees

BANK OF AMERICA CORPORATION

Dates

Publication Date: 20260505
Application Date: 20240813

Claims (15)

1 . A system for image and video authenticity verification, the system comprising: at least one non-transitory storage device comprising computer-readable code; and at least one processing device coupled to the at least one non-transitory storage device, wherein the computer-readable code is configured to cause the at least one processing device to: receive a submitted data file from a user device or entity system; identify if the submitted data file contains associated metadata; perform a first-pass metadata analysis and determine if the associated metadata of the submitted data file corroborates one or more known attributes of the submitted data file; generate an initial determination of inconclusive based on the first-pass metadata analysis; select one or more machine learning models of a machine learning dataset based on a file type of the submitted data file; perform one or more pattern recognition processes on the submitted data file via the identified one or more machine learning models; generate a determination of authenticity or inauthenticity of the submitted data file based on results of the one or more pattern recognition processes; generate a unique hash value for the submitted data file based on determining authenticity of the submitted data file; append the unique hash value to the metadata of the submitted data file; receive the submitted data file via the secure user application on the user device; based on receiving one or more user security verifications contemporaneous with creation of the submitted data file, automatically verify the submitted data file as authentic; overlay a visual logo on the data file based on determining authenticity of the submitted data file, thereby generating a visually authenticated data file; and transmit the visually authenticated data file to the user device.
2 . The system of claim 1 , wherein the submitted data file is a video file or an image file.
3 . The system of claim 1 , wherein the associated metadata further comprises exchangeable image file (EXIF) data.
4 . The system of claim 1 , wherein generating a determination of the authenticity or inauthenticity of the submitted data file further comprises: flagging characteristics in the submitted data file that indicate an inauthentic data point, comprising non-uniform pixel data or coloration; generating a confidence score representing a percentage confidence of the authenticity or inauthenticity of the submitted data file; and based on the percentage confidence of the authenticity or inauthenticity, converting the percentage confidence to a qualitative output of likely tampered with, cannot verify, or likely authentic.
5 . The system of claim 1 , wherein the computer-readable code is further configured to cause the at least one processing device to: upload the submitted data file to a remotely accessible datastore based on determining authenticity of the submitted data file; and transmit a link to the submitted data file to the user device.
6 . A computer program product for image and video authenticity verification, the computer program product comprising at least one non-transitory computer-readable medium having computer-readable program code portions embodied therein, the computer-readable program code portions comprising: an executable portion configured to receive a submitted data file from a user device or entity system; an executable portion configured to identify if the submitted data file contains associated metadata; an executable portion configured to perform a first-pass metadata analysis to determine if the associated metadata of the submitted data file corroborates one or more known attributes of the submitted data file; an executable portion configured to generate an initial determination of inconclusive based on the first-pass metadata analysis; an executable portion configured to select one or more machine learning models of a machine learning dataset based on a file type of the submitted data file; an executable portion configured to perform one or more pattern recognition processes on the submitted data file via the identified one or more machine learning models; an executable portion configured to generate a determination of authenticity or inauthenticity of the submitted data file based on results of the one or more pattern recognition processes; an executable portion configured to generate a unique hash value for the submitted data file based on determining authenticity of the submitted data file; an executable portion configured to append the unique hash value to the metadata of the submitted data file; an executable portion configured to receive the submitted data file via the secure user application on the user device; based on receiving one or more user security verifications contemporaneous with creation of the submitted data file, an executable portion configured to automatically verify the submitted data file as authentic; an executable portion configured to overlay a visual logo on the data file based on determining authenticity of the submitted data file, thereby generating a visually authenticated data file; and an executable portion configured to transmit the visually authenticated data file to the user device.
7 . The computer program product of claim 6 , wherein the submitted data file is a video file or an image file.
8 . The computer program product of claim 6 , wherein the associated metadata further comprises exchangeable image file (EXIF) data.
9 . The computer program product of claim 6 , wherein generating a determination of the authenticity or inauthenticity of the submitted data file further comprises: flagging characteristics in the submitted data file that indicate an inauthentic data point, comprising non-uniform pixel data or coloration; generating a confidence score representing a percentage confidence of the authenticity or inauthenticity of the submitted data file; and based on the percentage confidence of the authenticity or inauthenticity, converting the percentage confidence to a qualitative output of likely tampered with, cannot verify, or likely authentic.
10 . The computer program product of claim 6 further comprising: an executable portion configured to upload the submitted data file to a remotely accessible datastore based on determining authenticity of the submitted data file; and an executable portion configured to transmit a link to the submitted data file to the user device.
11 . A computer-implemented method for image and video authenticity verification, the method comprising: providing a computing system comprising a computer processing device and a non-transitory computer readable medium, wherein the computer readable medium comprises configured computer program instruction code, such that when said instruction code is operated by said computer processing device, said computer processing device performs the following operations: receive a submitted data file from a user device or entity system; identify if the submitted data file contains associated metadata; perform a first-pass metadata analysis and determine if the associated metadata of the submitted data file corroborates one or more known attributes of the submitted data file; generate an initial determination of inconclusive based on the first-pass metadata analysis; select one or more machine learning models of a machine learning dataset based on a file type of the submitted data file; perform one or more pattern recognition processes on the submitted data file via the identified one or more machine learning models; generate a determination of authenticity or inauthenticity of the submitted data file based on results of the one or more pattern recognition processes; generate a unique hash value for the submitted data file based on determining authenticity of the submitted data file; append the unique hash value to the metadata of the submitted data file; receive the submitted data file via the secure user application on the user device; based on receiving one or more user security verifications contemporaneous with creation of the submitted data file, automatically verify the submitted data file as authentic; overlay a visual logo on the data file based on determining authenticity of the submitted data file, thereby generating a visually authenticated data file; and transmit the visually authenticated data file to the user device.
12 . The computer-implemented method of claim 11 , wherein the submitted data file is a video file or an image file.
13 . The computer-implemented method of claim 11 , wherein the associated metadata further comprises exchangeable image file (EXIF) data.
14 . The computer-implemented method of claim 11 , wherein generating a determination of the authenticity or inauthenticity of the submitted data file further comprises: flagging characteristics in the submitted data file that indicate an inauthentic data point, comprising non-uniform pixel data or coloration; generating a confidence score representing a percentage confidence of the authenticity or inauthenticity of the submitted data file; and based on the percentage confidence of the authenticity or inauthenticity, converting the percentage confidence to a qualitative output of likely tampered with, cannot verify, or likely authentic.
15 . The computer-implemented method of claim 11 , further configured to: upload the submitted data file to a remotely accessible datastore based on determining authenticity of the submitted data file; and transmit a link to the submitted data file to the user device.

Description

CROSS-REFERENCE TO RELATED APPLICATION The present application claims priority to and is a continuation of U.S. patent application Ser. No. 17/343,103 filed Jun. 9, 2021, and of the same title; the contents of which are also incorporated herein by reference. BACKGROUND Given increased access to personal computing devices with requisite processing power, the advent of inauthentic imagery and audio is becoming more prevalent. For instance, it is possible for a consumer level device to execute an algorithm which can convincingly emulate the voice, likeness, or image of a human user without that user's knowledge, or without the viewer's awareness. As such, there is a need for systems and methods to identify and prevent inauthentic works, as well as quickly and easily convey a trusted visual indication of authenticity. BRIEF SUMMARY The following presents a simplified summary of one or more embodiments of the invention in order to provide a basic understanding of such embodiments. This summary is not an extensive overview of all contemplated embodiments, and is intended to neither identify key or critical elements of all embodiments, nor delineate the scope of any or all embodiments. Its sole purpose is to present some concepts of one or more embodiments in a simplified form as a prelude to the more detailed description that is presented later. The systems and methods described herein address the above needs by providing intelligent system and methods for identifying inauthentic works and providing a trusted source of verified authentic works. The system is adaptive, in that it can be adjusted based on the needs or goals of the user utilizing it, or may intelligently and proactively adapt based on the files or data received for processing. The system may be seamlessly embedded within existing applications or programs that the user may already use to interact with one or more entities, particularly those which aid in the managing of user resources. For instance, the system may be offered as an additional service or feature in an existing mobile application hosted by a trusted entity, or the like. The system may utilize various information in order to intelligently generate feedback regarding the authenticity of submitted works or data. Additionally, previously submitted works may be indicated by the system as authentic in a manner that the user can quickly and easily visually verify. Embodiments of the invention relate to systems, methods, and computer program products for image and video authenticity verification, the invention generally comprising at least one non-transitory storage device; and at least one processing device coupled to the at least one non-transitory storage device, wherein the at least one processing device is configured to: receive a submitted data file from a user device or entity system; identify if the submitted data file contains associated metadata; perform a first-pass metadata analysis and determine if the associated metadata of the submitted data file corroborates one or more known attributes of the submitted data file; generate an initial determination of authentic or inconclusive based on the first-pass metadata analysis; if the initial determination is inconclusive, identify one or more machine learning models of a machine learning dataset based on a file type of the submitted data file; and perform one or more pattern recognition processes on the submitted data file via the identified one or more machine learning models; and generate a determination of authenticity or inauthenticity of the submitted data file based on results of the one or more pattern recognition processes. In some embodiments, the submitted data file may be a video file or an image file. In some embodiments, the associated metadata further comprises exchangeable image file (EXIF) data. In some embodiments, the invention is further configured to: generate a unique hash value for the submitted data file based on determining authenticity of the submitted data file; append the unique hash value to the metadata of the submitted data file; and transmit a notification of authenticity of the submitted data file to the user device. In some embodiments, the invention is further configured to: upload the submitted data file to a remotely accessible datastore based on determining authenticity of the submitted data file; and transmit a link to the submitted data file to the user device. In some embodiments, the invention is further configured to: overlay a visual logo on the data file based on determining authenticity of the submitted data file, generating a visually authenticated data file; and transmit the visually authenticated data file to the user device. In some embodiments, the invention is further configured to: receive the submitted data file via a secure user application on the user device; and based on receiving one or more user security verifications contemporaneous with creation of the submitted data file, autom