ATM False Alarm Prevention via Image Template Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing self-service terminals, such as ATMs, face issues with false alarms due to variations in lighting conditions and reflections, making it difficult to distinguish between legitimate and illegitimate manipulations, such as skimming device installations, which can lead to unnecessary alerts and operational disruptions.
Innovation Solution
A device and method that utilize camera templates stored on the self-service machine to compare alarm images with pre-defined templates, using techniques like cross-correlation and normalized Euclidean distance, to differentiate between legitimate and false alarms, thereby reducing false alerts and allowing for localized reaction to specific circumstances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If camera surveillance systems are used to detect tampering attempts, then the ability to detect manipulations is improved, but the number of false alarms increases due to lighting variations and reflections
Solution Approach 1:
The system performs preliminary action by capturing reference images under normal operating conditions and storing them as templates before actual tampering events occur. These templates represent the baseline appearance of the control panel under various lighting conditions, enabling subsequent comparisons to distinguish between normal variations and actual tampering attempts.
Solution Approach 2:
The system creates copies of the control panel's appearance under different lighting conditions by capturing reference images and storing them as templates. These template copies are then used for comparison during actual monitoring, allowing the system to differentiate between legitimate visual variations and genuine tampering attempts without requiring complex real-time analysis.
2Area of stationary object
If multiple cameras are mounted to capture different areas, then the coverage of the monitoring system is improved, but the complexity of the system increases
Solution Approach 1:
The system segments the monitoring task by assigning different cameras to capture specific areas of interest (control panel, user area, card slot, cash dispenser). Each camera focuses on a particular zone, and the system processes images from each camera independently using dedicated template comparisons, reducing the complexity of analyzing all areas simultaneously while maintaining comprehensive coverage.
Data Source
Figure 1
Figure 2
AI summary
A method for preventing false alarms on a self-service machine, in particular an ATM, wherein the self-service machine has at least one camera and at least one alarm unit that generates alarms when the self-service machine is attacked, and has a network connection to a network to forward the alarms to a customer, comprising the steps of: - generating an alarm, by the alarm unit, and storing alarm images from the camera from which the alarm was derived; - before forwarding the alarm via a digital network to an operator, comparing the digital alarm images with templates stored on the self-service machine that indicate a false alarm, and if a match with the templates is found, the alarm is ignored; otherwise, the alarm is forwarded via the network to the customer.