Live masking
AssigneeU.S. Bank National Association
InventorsSathiskumar Meganathan, Rajasekharam Talluri
Life of this patent
Estimating the term…
Abstract
Various embodiments are directed to apparatuses, methods, computer-readable media, computer program products, and systems related to establishing a first connection with a user component; establishing a second connection with an agent component; receiving, from the user component and via the first connection, an image representative of at least a portion of a graphical user interface, the portion of the graphical user interface comprising sensitive text; applying the image to a masking engine to generate a masked image by masking at least the sensitive text; and transmitting, to the agent component and via the second connection, the masked image.
Claims23
A system comprising:
- a first client device comprising a user component, the user component configured to cause capture of images representative of at least a portion of a graphical user interface displayed at the first client device;
- a second client device comprising an agent component;
- and
- at least one secure server comprising at least one processor, at least one communications circuitry, and at least one memory, the at least one memory comprising computer coded instructions therein, wherein the computer coded instructions are configured to, when executed by the at least one processor, cause the at least one secure server to:
- establish, via the at least one communications circuitry, a first connection between the user component and the at least one secure server;
- establish, via the at least one communications circuitry, a second connection between the agent component and the at least one secure server;
- receive, from the first client device via the first connection, an image representative of at least a portion of the graphical user interface displayed at the first client device;
- apply, via the at least one secure server, the image to a masking engine by:
- extracting, via an optical character recognition (OCR) model, text of the graphical user interface from the image to generate extracted text;
- classifying, via a text-based classification model applied to the extracted text, at least a portion of the extracted text as sensitive text based on contents of the extracted text;
- and
- masking, via an image processing model, the sensitive text to generate a masked image;
- and
- transmit, to the agent component via the second connection, at least one masked image comprising the masked image.
A computer-implemented method comprising:
- establishing, via one or more processors, a first connection with a user component, the user component configured to cause capture of images representative of at least a portion of a graphical user interface displayed at a client device;
- establishing, via one or more processors, a second connection with an agent component;
- receiving, from the user component via the first connection, an image representative of at least a portion of the graphical user interface, the portion of the graphical user interface comprising sensitive text;
- applying the image to a masking engine by:
- extracting, via an optical character recognition (OCR) model, text of the graphical user interface from the image to generate extracted text;
- classifying, via a text-based classification model applied to the extracted text, at least a portion of the extracted text as sensitive text based on contents of the extracted text;
- and
- masking, via an image processing model, the sensitive text to generate a masked image;
- and
- transmitting, via one or more processors, to the agent component via the second connection, the masked image.
A server comprising at least one processor and at least one memory, the at least one memory comprising computer coded instructions therein, wherein the computer coded instructions are configured to, when executed by the at least one processor, cause the server to:
- establish a first connection with a user component, the user component configured to cause capture of images representative of at least a portion of a graphical user interface displayed at a client device;
- establish a second connection with an agent component;
- receive, from the user component via the first connection, an image representative of at least a portion of the graphical user interface, the portion of the graphical user interface comprising sensitive text;
- apply the image to a masking engine by:
- extracting, via an optical character recognition (OCR) model, text of the graphical user interface from the image to generate extracted text;
- classifying, via a text-based classification model applied to the extracted text, at least a portion of the extracted text as sensitive text based on contents of the extracted text;
- and
- masking, via an image processing model, the sensitive text to generate a masked image;
- and
- transmit, to the agent component via the second connection, the masked image.
Dependencies are inferred from the claim text.