Counter-fraud measures for an ATM device

An automated teller machine (ATM) device, comprising:

• one or more memories;
• and
• one or more processors communicatively coupled to the one or more memories, configured to:
• monitor a target area associated with the ATM device by capturing images of the target area using one or more image sensors,
• the target area being associated with potentially fraudulent activity targeting the ATM device;
• perform, using a machine learning application that is trained to identify human body parts, image segmentation of one or more of the images to detect one or more conditions,
• wherein the one or more conditions are associated with an indication of fraudulent activity, and
• wherein the machine learning application is a deep learning network that is trained to determine specific hand motions indicative of fraudulent activity associated with the ATM device;
• detect the one or more conditions;
• generate, based on detecting the one or more conditions, a fraud score associated with the one or more conditions;
• determine whether the fraud score meets a threshold;
• and
• perform, based on whether the fraud score meets the threshold, one or more actions.

A non-transitory computer-readable medium storing instructions, the instructions comprising:

• one or more instructions that, when executed by one or more processors, cause the one or more processors of an automated teller machine (ATM) device to:
• receive a request to process an ATM transaction;
• dispense, via an instrument dispenser output, a plurality of instruments based on the request;
• perform, using a machine learning application that is trained to identify human body parts, image segmentation of one or more images to detect one or more conditions,
• wherein the one or more conditions are associated with an indication of fraudulent activity, and
• wherein the machine learning application is a deep learning network that is trained to perform image segmentation of human body parts by analyzing sequences of images in which fraudulent activity occurs;
• detect, based on performing the image segmentation, that a user's hand approaches the instrument dispenser output after dispensing the plurality of instruments;
• determine, after dispensing the plurality of instruments and after detecting the user's hand approaches the instrument dispenser output, that a portion of the plurality of instruments is present at the instrument dispenser output;
• and
• perform one or more actions based on determining that a portion of the plurality of instruments is present at the instrument dispenser output.