METHOD AND DEVICE FOR AVOIDING FALSE ALARMS IN MONITORING SYSTEMS

• US 20150206422A1
• Filed: 01/14/2015
• Published: 07/23/2015
• Est. Priority Date: 01/17/2014
• Status: Active Grant

• Current Assignee: Wincor Nixdorf International GmbH (Diebold Nixdorf, Incorporated)
• Original Assignee: Wincor Nixdorf International GmbH (Diebold Nixdorf, Incorporated)
• Inventors: Priesterjahn, Steffen, Drichel, Alexander
• Granted Patent: US 9,870,700 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G07F 19/207 Surveillance aspects at ATMs
• CPC Class Codes: G08B 13/19695 Arrangements wherein non-video detectors start video recording or forwarding but do not generate an alarm themselves
• CPC Class Codes: G08B 29/185 Signal analysis techniques for reducing or preventing false alarms or for enhancing the reliability of the system
• CPC Class Codes: H04N 7/183 for receiving images from a single remote source

Abstract

A method is provided for avoiding false alarms on a self-service terminal, in particular an ATM. The ATM has at least one camera and at least one alarm unit that generates alarms when the ATM is attacked, and a network connection to a network to pass on the alarms to a customer. The method includes generating an alarm by the alarm unit and storing alarm images from the camera from which the alarm was derived; before passing on the alarm to an operator via a digital network, comparing the digital alarm images with templates stored on the ATM and that characterise a false alarm, and if a correlation with the templates is established, the alarm is ignored; otherwise the alarm is passed on to the consumer by the network.

Claims

1. A method for avoiding false alarms on a self-service terminal, being an ATM, wherein the ATM has at least one camera and at least one alarm unit that generates alarms when the ATM is attacked, and a network connection to a network in order to pass on the alarms to a customer, comprising the steps:

generating an alarm by the alarm unit and storage of alarm images from the camera from which the alarm was derived;

before passing on the alarm to an operator via a digital network, comparing the digital alarm images with templates that are stored on the ATM and that characterise a false alarm, and ignoring the alarm if a correlation with the templates is established, otherwise passing the alarm on to the consumer by the network.

2. The method of claim 1, wherein after reception of the alarm by the network and after analysis of the alarm over the network, the method further comprising triggering a template generation on the ATM if the customer is of the opinion that this is a false alarm that is to be ignored in the future, wherein then the alarm images on the ATM are converted into templates in order to avoid a similar alarm in the future.

3. The method of claim 1, wherein comparison of the digital alarm images with templates takes place by one or more of the following methods:

cross correlation, CC

normalised cross correlation, NCC,

squared Euclidean distance, SED,

normalised square Euclidean distance, NSED.

4. The method of claim 3, wherein for NSED, a blurring of the images is applied, in that a Gaussian filter or other filter is used.

5. The method of claim 1, wherein the template consists of image data and metadata, wherein the metadata comprise one or more of the following items of information:

a) information from preprocessed images

aperture

b) camera information

serial number

calibration at the area of interest

camera measurements

c) system information of the ATM

serial number

model description,

wherein the metadata and image data are stored in a file and a template is defined, and wherein the metadata also influence the comparison.

6. The method of claim 1, wherein the templates comprise several images or image series, or videos that are compared with alarm image series.

7. The method of claim 6, wherein the templates with several images manage images from different cameras.

8. The method of claim 1, wherein the ATM is switched into an automatic template mode for calibration of the camera and for storage of the metadata for the camera.

9. A self-service ATM, comprising a camera, an alarm unit that generates alarms when the ATM is attacked, a network connection to a network for passing on the alarms to a customer, with a device for avoiding false alarms that is configured to compare alarm images, generated in the case of an alarm, from the camera from which the alarm was derived, with templates stored on the ATM and that characterise a false alarm, and if a correlation with the templates can be established, to ignore the alarm;

otherwise to pass on the alarm to a customer via the network connection and the network.

10. The ATM of claim 9, wherein the device for avoiding false alarms is configured, on instruction, following reception of the alarm via the network and after an analysis of the alarm over the network, to generate a template on the ATM if the customer is of the opinion that this is a false alarm that is to be ignored in the future, in order then to convert the alarm images on the ATM into templates to avoid a similar alarm in the future.

11. The ATM of claim 10, wherein the device for avoiding false alarms is configured to compare the digital alarm images with templates, using one or more of:

cross correlation, CC,

normalised cross correlation, NCC,

squared Euclidean distance, SED,

normalised squared Euclidean distance, NSED.

12. The ATM of claim 10, wherein the device for avoiding false alarms is configured, for NSED, to apply a blurring of the images in which a filter, a Gaussian filter or other filter, can be used.

13. The ATM of claim 9, wherein the template consists of image data and metadata, wherein the metadata comprise one or more of the following items of information:

a) information from the preprocessed images

aperture

b) camera information

serial number

calibration at the area of interest

camera measurements

c) system information of the ATM

serial number

model description,

and wherein the metadata and image data are stored in a file and a template is defined, and wherein the device for avoiding false alarms is configured to use the metadata to influence the comparison.

14. The ATM of claim 9, wherein the templates comprise several images, image series of videos that can be compared with alarm image series,

and/or wherein several images from different cameras are stored in the templates.

15. The ATM of claim 9, wherein the ATM can be switched into automatic template mode for calibration of the camera and for storage of the metadata for the camera.