Artificial intelligence (AI)-based security systems for monitoring and securing physical locations

Patent metadata

Field	Value
Publication number	US-11735017-B2
Application number	US-202117356032-A
Country	US
Kind code	B2
Filing date	Jun 23, 2021
Priority date	Jun 23, 2021
Publication date	Aug 22, 2023
Grant date	Aug 22, 2023

Abstract

Various aspects of the disclosure relate to monitoring a physical location to determine and/or predict anomalous activities. One or more machine learning algorithms may be used to analyze inputs from one or more sensors, cameras, audio recording equipment, and/or any other types of sensors to detect anomalous measurements/patterns. Notifications may be sent one or more devices in a network based on the detection.

First claim

The invention claimed is:

1. An artificial intelligence (AI)-based system for monitoring and securing a retail banking physical location based on millimeter wave scanner measurements, the system comprising:
a millimeter wave scanner configured to scan individuals, at the physical location, for detecting concealed objects;
one or more actuators located at the physical location;
an electronic monitoring platform communicatively coupled to the millimeter wave scanner and the one or more actuators, the electronic monitoring platform comprising:
at least one processor;
a communication interface communicatively coupled to the at least one processor; and
memory storing computer-readable instructions that, when executed by the at least one processor, cause the electronic monitoring platform to:
receive, via the communication interface from the millimeter wave scanner, measurements corresponding to concealed objects associated with a plurality of individuals at the physical location;
perform cluster analysis on the measurements to determine groups of normal measurements corresponding to the concealed objects associated with the plurality of individuals;
receive, from the millimeter wave scanner and for an individual at the physical location, a measurement corresponding to concealed objects associated with the individual;
determine, based on the groups of normal measurements, that the measurement corresponding to concealed objects associated with the individual is anomalous; and
based on the determining:
send, via the communication interface, one or more notifications to the one or more actuators, wherein at least one actuator of the one or more actuators is associated with a shutter at a teller window in the physical location, and wherein the actuator causes the shutter to activate based on receiving a notification; and
send, via the communication interface and to a computing device,

an alert notification.

2. The system of claim 1 , wherein at least one second actuator of the one or more actuators is associated with a vault in the physical location, and wherein the second actuator causes the vault to lock based on receiving a second notification.

3. The system of claim 1 , wherein the measurements correspond to concealed metallic objects.

4. The system of claim 1 , wherein the performing the cluster analysis comprises performing one of:
hierarchical clustering;
centroid based clustering;
density based clustering;
distribution based clustering; and
combinations thereof.

5. The system of claim 1 , wherein the determining that the measurement is anomalous is based on determining that distances between the measurement and core points associated with the groups are greater than a threshold value.

6. The system of claim 1 , wherein the system further comprises:
one or more cameras configured to capture images associated with the physical location; and
wherein the computer-readable instructions when executed by the at least one processor, cause the electronic monitoring platform to:
receive, via the communication interface and from the one or more cameras, the images; and
perform, based on a convolutional neural network, image recognition analysis of the images to determine that the individual was at the physical location an anomalous number of times;
wherein the sending the one or more notifications is further based on the determining that the individual was at the physical location an anomalous number of times.

7. The system of claim 1 , wherein the system further comprises:
one or more cameras configured to capture videos associated with the physical location; and
wherein the computer-readable instructions when executed by the at least one processor, cause the electronic monitoring platform to:
receive, from the one or more cameras, the videos;
determine, based on the videos, gaits, paces, and postures associated with the plurality of individuals, wherein the performing the cluster analysis to determine the groups is further based on the determined gaits, paces, and postures; and
determine, based on the videos, a gait, a pace, and a posture of the individual, wherein the determining that the measurement corresponding to concealed objects associated with the individual is anomalous is further based on the gait, the pace, and the posture of the individual.

8. The system of claim 7 , wherein the computer-readable instructions when executed by the at least one processor, cause the electronic monitoring platform to determine the gaits, the paces, and the postures associated with the plurality of individuals based on an AI model, wherein the AI model is based on one of:
a logistic regression model;
a decision tree model;
a random forest model;
a neural network;
a support vector machine; and
combinations thereof.

9. The system of claim 1 , wherein the measurements comprise one of:
sizes of the concealed objects;
locations of the concealed objects; and
combination thereof.

10. A method for remote monitoring and securing a retail banking physical location based on millimeter wave scanner measurements, the method comprising:
receiving, at an electronic monitoring platform and from a millimeter wave scanner, measurements corresponding to concealed objects associated with a plurality of individuals at the physical location;
performing cluster analysis on the measurements to determine groups of normal measurements corresponding to the concealed objects associated with the plurality of individuals;
receiving, from the millimeter wave scanner, a measurement corresponding to concealed objects associated with an individual at the physical location;
determining, based on the groups of normal measurements, that the measurement corresponding to concealed objects associated with the individual is anomalous; and
based on the determining:
sending one or more notifications to one or more actuators, wherein at least one actuator of the one or more actuators is associated with a shutter at a teller window in the physical location, and wherein the actuator causes the shutter to activate based on receiving a notification; and
sending, to a computing device, an alert notification.

11. The method of claim 10 , wherein at least one second actuator of the one or more actuators is associated with a vault in the physical location, and wherein the second actuator causes the vault to lock based on receiving a second notification.

12. The method of claim 10 , wherein the measurements correspond to concealed metallic objects.

13. The method of claim 10 , wherein the performing the cluster analysis comprises performing one of:
hierarchical clustering;
centroid based clustering;
density based clustering;
distribution based clustering; and
combinations thereof.

14. The method of claim 10 , wherein the determining that the measurement is anomalous is based on determining that distances between the measurement and core points associated with the groups are greater than a threshold value.

15. The method of claim 10 , further comprising:
receiving, from one or more cameras at the physical location, images associated with the physical location; and
performing, based on a convolutional neural network, image recognition analysis of the images to determine that the individual was at the physical location an anomalous number of times;
wherein the sending the one or more notifications is further based on the determining that the individual was at the physical location an anomalous number of times.

16. The method of claim 10 , further comprising:
receiving, from one or m

Assignees

• Bank Of America

Inventors

• Albero George Anthony
• Conley Nathalie Elizabeth
• Knight Elizabeth Jordan
• Nguyen-Lo Lauren Thanh
• Armstrong Robert Jon

Classifications

• G08B13/19613 Primary

  Recognition of a predetermined image pattern or behaviour pattern indicating theft or intrusion · CPC title

• G06F18/23

  Clustering techniques · CPC title

• G06N3/02

  Neural networks · CPC title

• G06N20/10

  using kernel methods, e.g. support vector machines [SVM] · CPC title

• G08B13/1609

  using active vibration detection systems · CPC title