Assessing the security situation by means of IoT activity sensors

Patent metadata

Field	Value
Publication number	US-10997834-B2
Application number	US-201916391787-A
Country	US
Kind code	B2
Filing date	Apr 23, 2019
Priority date	Apr 27, 2018
Publication date	May 4, 2021
Grant date	May 4, 2021

Abstract

A method and/or system for assessing the security situation in a building through an evaluation of sensor values provided by activity sensors situated in the building, in particular, in the accommodation region, an estimate of the number of persons actually present in the building, determining the number of persons to be expected in the building by an evaluation of administrative data (HR, Outlook, events, occupancies, etc.). Based upon a comparison of the number of persons to be expected with the number of actual persons, an indicator is determined for assessing the actual security situation in the building or in the accommodation region.

First claim

The invention claimed is:

1. A method for assessing the security situation in a building, the method comprising:
evaluating sensor values provided by activity sensors located in the building to estimate a number of persons actually present in the building;
evaluating a number of persons expected to be present in the building based on expected activity values;
comparing the number of persons to be expected to the number of persons actually present to assess the actual security situation in the building; and
wherein the expected activity values include occupancy data based on an electronic calendar system;
if the actual security situation in the building includes a disparity between the number of persons to be expected and the number of persons actually present exceeding a predefined threshold value, initiating automatic security measures including zooming and/or panning a camera.

2. The method as claimed in claim 1 , wherein:
the activity sensors are connected via an IP communication interface to a Cloud Service Application in a Cloud infrastructure;
the Cloud Service Application determines the estimate of the number of persons actually present in the building or the accommodation region;
the Cloud Service Application determines the number of persons to be expected; and
the Cloud Service Application compares the number of persons to be expected to the number of persons actually present.

3. The method as claimed in claim 1 , wherein the estimate of the number of persons actually present in the building is determined using artificial intelligence.

4. The method as claimed in claim 3 , wherein estimating the number of persons actually present in the building includes using Supervised Learning.

5. A system for assessing the security situation in an accommodation region of a building, the system comprising:
activity sensors mounted in an accommodation region of the building, wherein the activity sensors provide sensor values via a respective IP interface to a server;
the server configured to:
receive and store the sensor values from the activity sensors and determine an estimate of the number of persons actually present in the accommodation region on the basis of the sensor values;
determine an expected number of persons in the accommodation region by evaluating expected activity values associated with the accommodation region;
assess an actual security situation in the accommodation region by comparing the number of the persons to be expected with the number of estimated actual persons; and
if the actual security situation in the building includes a disparity between the number of persons to be expected and the number of persons actually present exceeding a predefined threshold value, initiating automatic security measures including zooming and/or panning a camera

wherein the expected activity values include occupancy data based on an electronic calendar system.

6. The system as claimed in claim 5 , wherein the activity sensors comprise IoT devices connected via an IP network to the server.

7. The system as claimed in claim 5 , wherein:
the server comprises an AI engine used to determine the estimate of the number of persons actually present in the accommodation region; and
the AI engine employs methods of Deep Learning and/or Supervised Learning.

8. The system as claimed in claim 5 , wherein the server comprises a Cloud infrastructure.

9. A Cloud Service Application for assessing a security situation in an accommodation region of a building, wherein the Cloud Service Application is stored on a non-transitory memory and configured, when executed by a processor, causes the processor to:
evaluate sensor values provided by IoT activity sensors situated in the accommodation region;
estimate a number of persons actually present in the accommodation region;
determine a number of persons expected to be in the accommodation region by evaluating expected activity values related to the accommodation region;
to assess the security situation based upon a comparison of the number of persons to be expected with the number of actual persons;
wherein the expected activity values include occupancy data based on an electronic calendar system; and
if the actual security situation in the building includes a disparity between the number of persons to be expected and the number of persons actually present exceeding a predefined threshold value, initiating automatic security measures including zooming and/or panning a camera.

10. The Cloud Service Application as claimed in claim 9 , wherein the Cloud Service Application is connected via an IP network to the IoT activity sensors.

Assignees

• Siemens Schweiz Ag

Inventors

• Domeisen Patrik
• Konrad Hilmar

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06Q10/00 Primary

  Administration; Management · CPC title

• G08B13/20

  Actuation by change of fluid pressure · CPC title

• G08B26/007

  Wireless interrogation · CPC title

• G06N3/08

  Learning methods · CPC title