Apparatus, method, and non-transitory computer-readable medium for determining automatic teller machine (ATM) incident

1 . An information processing apparatus comprising:

at least one memory storing instructions; and

at least one processor configured to execute the instructions to:

determine whether a user who has visited an automatic teller machine (ATM) is talking on a phone based on a comparison between skeleton information of the user and registered skeleton information, the skeleton information being obtained from first data by machine learning, the first data being a captured image related to a target area based on the ATM, and the registered skeleton information being information related to a phone call action;

start acquisition of second data that suggests a state of the user and has a data type different from that of the first data in a case where it is determined that the user is talking on a phone;

acquire a determination result of presence or absence of an incident using the first data and the second data;

update, based on the determination result of presence or absence of the incident, the registered skeleton information;

switch an operation mode of the ATM from a first operation mode, for executing a normal operation, to a second operation mode, for outputting information at least partially different from information outputted in the first operation mode, in a case where it is determined that the user is talking on a phone;

transmit data regarding a state of the user in a target period based on the switching to the second operation mode to an interested party terminal and acquire the determination result of presence or absence of the incident from a related party terminal; and

change an operation content of the second operation based on at least the determination result of presence or absence of the incident.

2 . The information processing apparatus according to claim 1 , wherein at least one processor further configured to execute the instructions to:

estimate a talk content based on a voice data acquired at a predetermined position with respect to the ATM, which is the second data; and

determine presence or absence of the incident based on the captured image and the talk content.

3 . The information processing apparatus according to claim 1 , wherein at least one processor configured to execute the instructions to:

transmit the first data and the second data to an interested party terminal used by an interested party; and

receive, from the interested party terminal, a determination result of presence or absence of an incident input to the interested party terminal.

4 . The information processing apparatus according to claim 3 , wherein at least one processor configured to execute the instructions to transmit user information regarding the user to the interested party terminal together with the first data and the second data.

5 . The information processing apparatus according to claim 1 , wherein at least one processor configured to execute the instructions to cause the ATM to output question information stored in advance in an operation control database (DB) in the second operation mode.

6 . The information processing apparatus according to claim 5 , wherein at least one processor configured to execute the instructions to acquire the determination result of the presence or absence of the incident by further using a response to the question information.

7 . The information processing apparatus according to claim 5 , wherein at least one processor configured to execute the instructions to:

perform primary determination on the presence or absence of the incident based on a response to the question information;

transmit the first data and the second data to an interested party terminal used by an interested party when a result of the primary determination indicates the presence of the incident; and

receive, from the interested party terminal, a determination result of presence or absence of an incident input to the interested party terminal.

8 . The information processing apparatus according to claim 1 , wherein the at least one processor is configured to execute the instructions to determine whether the user is talking on the phone based on a similarity between the skeleton information about a predetermined region of a body and the registered skeleton information about the predetermined region.

9 . An information processing method comprising:

determining, by a computer, whether a user who has visited an automatic teller machine (ATM) is talking on a phone based on a comparison between skeleton information of the user and registered skeleton information, the skeleton information being obtained from first data by machine learning, the first data being a captured image related to a target area based on the ATM, and the registered skeleton information being information related to a phone call action;

starting, by the computer, acquisition of second data that suggests a state of the user and has a data type different from that of the first data in a case where it is determined that the user is talking on a phone;

acquiring, by the computer, a determination result of presence or absence of an incident using the first data and the second data;

updating, by the computer, based on the determination result of presence or absence of the incident, the registered skeleton information;

switching an operation mode of the ATM from a first operation mode, for executing a normal operation, to a second operation mode, for outputting information at least partially different from information outputted in the first operation mode, in a case where it is determined that the user is talking on a phone;

transmitting data regarding a state of the user in a target period based on the switching to the second operation mode to an interested party terminal and acquire the determination result of presence or absence of the incident from a related party terminal; and

changing an operation content of the second operation based on at least the determination result of presence or absence of the incident.

10 . A non-transitory computer-readable medium storing a program for causing a computer to execute:

a procedure comprising determining whether a user who has visited an automatic teller machine (ATM) is talking on a phone based on a comparison between skeleton information of the user and registered skeleton information, the skeleton information being obtained from first data by machine learning, the first data being a captured image related to a target area based on the ATM, and the registered skeleton information being information related to a phone call action;

starting acquisition of second data that suggests a state of the user and has a data type different from that of the first data in a case where it is determined that the user is talking on a phone;

acquiring a determination result of presence or absence of an incident using the first data and the second data; and

updating, based on the determination result of presence or absence of the incident, the registered skeleton information;

switching an operation mode of the ATM from a first operation mode, for executing a normal operation, to a second operation mode, for outputting information at least partially different from information outputted in the first operation mode, in a case where it is determined that the user is talking on a phone;

transmitting data regarding a state of the user in a target period based on the switching to the second operation mode to an interested party terminal and acquire the determination result of presence or absence of the incident from a related party terminal; and

changing an operation content of the second operation based on at least the determination result of presence or absence of the incident.