ATM Card Skimmer Detection Using Wavelet Transform and Neural Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for detecting card skimmers at ATMs suffer from low accuracy and frequent false positive alerts, particularly due to issues with capacitive sensor signals caused by weather conditions like rain.
Innovation Solution
A computer-implemented method using wavelet transforms and machine learning models to differentiate between card skimmer alerts and weather-related noise, generating ambient environment alerts to reduce false positives and enhance detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If capacitive sensors are used to detect card skimmers at ATMs, then detection capability is provided, but false positive alerts occur due to weather conditions like rain
Solution Approach 1:
The patent segments the detection signal into different frequency components using wavelet transform. By dividing the complex sensor signal into multiple frequency bands, the system can separately analyze and identify rain-induced signals versus skimmer-related signals, thereby reducing false positives while maintaining detection accuracy
Solution Approach 2:
The patent transforms the detection problem from the time domain to the frequency domain using wavelet transform. This dimensional change allows the system to analyze signals based on frequency characteristics rather than temporal patterns, enabling differentiation between rain noise and actual skimmer detections through spectral analysis
2Measurement precision
If wavelet transform and machine learning models are used to process detection data, then detection accuracy is improved, but data processing complexity increases
Solution Approach 1:
The patent applies wavelet transform as a preliminary processing step before machine learning classification. By pre-processing the raw sensor data to extract frequency-domain features and identify rain periods, the system reduces the complexity of the subsequent machine learning task, allowing for more accurate detection with manageable computational requirements
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method achieves high accuracy in card skimmer detection, minimizing false alerts and ensuring reliable ATM operation by distinguishing between skimmer activity and weather-induced noise, thus reducing unnecessary maintenance and maintaining customer confidence.
Implementation Method 1
The detection data may be converted from the time domain to the frequency domain using a wavelet transform to generate the wavelet matrix
Data Source
AI summary
A system for detecting mounting of a card skimmer at an automated teller machine receives detection data from sensors. The system converts the detection data from the time domain to the frequency domain using a wavelet transform to generate a wavelet matrix of coefficients. The system includes a convolutional neural network model to identify features in the wavelet matrix. The system generates a rain alert signal based on the identified features. The system includes a long short-term memory recurrent neural network model to determine a sequence of (i) an idle state, (ii) a hand approach state, and (iii) a card skimmer mounting state based on the detection data. The system generates a tamper alert signal based on the sequence determination.