Ncr Corporation

Ncr Voyix Corporation

Ncr Atleos Corporation

Conor Michael Fyfe

Alexander William Whytock

Graeme Mitchell

Shannon Mcgrath

A computer-implemented method and transaction terminal is disclosed. The method comprises capturing, by an imaging device, a first image of an internal region of an apparent peripheral device of the transaction terminal, calculating, by a processor of a transaction terminal, a first amount of change between the first image and a baseline image of the internal region, capturing, by the imaging device, at least one further image of the internal region, for each further image, calculating, by the processor, a second amount of change between the further image and a respective baseline image of the internal region, and when at least two consecutive values of the first and/or second amount of change are below a predefined value, determining that the imaging device is in a dummy peripheral device to prevent detection of a foreign object in an active peripheral device of the transaction terminal.

Detection of peripheral device manipulation

Images captured for components of a device are monitored for changes by evaluating a first region of interest in the images. Periodically, a command is sent to the device to move one or more of the components to a known position or state. A certain component or set of components associated with being moved based on the command is evaluated in a second region of interest in the images to determine if the corresponding component or set of components is in the known position or state within the images. When the corresponding component or set of components is not identified from the images in the known position or state, a security alert is raised for the device and security operations are processed on the host device.

Replay attack detection

Video of customer faces during checkouts are captured in real time and analyzed to assign identifiers for expressions of the customers. Simultaneously, transaction data and events being raised for the checkouts are correlated based on time stamps with the expression identifiers. A collection of data captured for the correlated transaction data, events, and identifiers are processed by a machine-learning algorithm to identify clusters and/or patterns between the expression identifiers and specific transaction events. The clusters are mined and reported to a retailer for identifying points during the workflow for the checkouts, transaction interface components, transaction terminal settings, transaction terminal peripherals, etc. that are potentially causing negative customer reactions during the checkouts.

Growing community of inventors

Carnoustie, United Kingdom

Conor Michael Fyfe