Banknote Validation Using Weighted Sensor Scores
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional banknote validation methods in Self Service Terminals, such as ATMs, often inaccurately determine the authenticity of banknotes due to the voting-based combination of sensor decisions, which disregards the confidence levels of individual sensors, leading to potential misclassification of banknotes as genuine, suspect, or counterfeit.
Innovation Solution
A method and apparatus that apply weighting factors to each sensor's validation score, using membership functions and fuzzy logic to generate a combined validation score, allowing for a more accurate determination of a banknote's authenticity by considering the importance of each sensor and introducing 'fuzziness' into the decision-making process, thereby distinguishing between 'genuine', 'suspect', and 'counterfeit' classifications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If voting-based combination method is used to combine sensor decisions, then the decision-making process is simple, but the accuracy of banknote validation deteriorates because confidence levels of individual sensors are disregarded
Solution Approach 1:
The patent changes the parameter of combining sensor decisions from simple voting (0/1 outcomes) to weighted averaging using validation scores (0-1 range) that incorporate confidence levels. This transforms the combination method to preserve both simplicity and accuracy by using normalized scores with weighting factors that reflect sensor reliability.
Solution Approach 2:
The patent replaces the mechanical voting system (where each sensor has equal weight) with a more sophisticated weighted scoring system that substitutes the rigid voting mechanism with a flexible calculation based on validation scores and weighting factors, enabling nuanced decision-making.
2Device complexity
If conventional voting method is used to combine sensor decisions, then the system complexity is low, but the reliability of validation decisions deteriorates due to inability to express sensor confidence
Solution Approach 1:
The patent introduces validation scores as a new parameter that ranges from 0 to 1, replacing the binary voting outcome. This new parameter allows the system to express confidence levels while maintaining reasonable complexity through standardized calculations using weighting factors and membership functions.
Solution Approach 2:
The patent makes the system dynamic by allowing weighting factors to be adjusted based on sensor performance and configuration. This enables the system to adapt to different operational conditions and maintain high reliability without requiring overly complex fixed structures.
3Ease of manufacture
If equal weighting is assigned to all sensors, then the system is simple to implement, but the validation accuracy deteriorates when sensors have different importance or reliability
Solution Approach 1:
The patent changes the weighting parameter from equal weight (1.0 for all sensors) to variable weighting factors that can be adjusted based on sensor reliability and importance. This allows the system to maintain implementation simplicity while achieving higher accuracy through configurable weighting.
Solution Approach 2:
The patent introduces dynamic weighting where the importance of each sensor can be adjusted based on operational conditions, sensor performance, and specific application requirements. This enables the system to optimize accuracy for different scenarios without requiring complete redesign.
Data Source
AI summary
A method of validating a media item, comprising: for each of a plurality of sensors that sense a media item, determining a respective validation score indicating a likelihood that the sensed media item is valid; and determining if the media item is valid responsive to the plurality of validation scores. Apparatus for validating a media item and a document processing module are also provided.