Automated Banknote Recognition via Segmentation Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional banknote recognition systems in self-service terminals are labor-intensive and require frequent updates due to the complexity of recognizing multiple denominations and orientations of banknotes, including counterfeit ones, which hampers efficient validation during bunch deposits.
Innovation Solution
The development of automated media-recognition templates using segmentation maps and feature sets, which allow for the creation of a one-class classifier for genuine banknotes, enabling efficient recognition and validation by analyzing statistical representations of sub-regions, reducing the need for human intervention and frequent updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional currency templates are used for banknote recognition, then recognition capability is achieved, but the process becomes extremely labor intensive and requires frequent updates
Solution Approach 1:
The system automatically creates templates by analyzing images of banknotes without requiring manual feature selection. The template generation process is self-service, where the system itself identifies and extracts key features from training images, eliminating the need for human experts to manually create and update templates for each banknote denomination and orientation.
Solution Approach 2:
The patent replaces the manual mechanical process of creating currency templates with an automated computational system. Instead of human operators manually selecting and marking key features on banknote images, the system uses image processing algorithms to automatically identify, extract, and store template features, substituting mechanical human labor with automated digital processing.
2Reliability
If traditional currency templates are updated for new banknote releases, then recognition accuracy is maintained, but human expertise and time are consumed
Solution Approach 1:
The system performs preliminary action by automatically analyzing and storing template features in advance before new banknote releases occur. The template database is pre-populated with features extracted from training images, so when new banknotes are introduced, the system can quickly generate new templates using the same automated process without requiring human experts to immediately update existing templates.
Solution Approach 2:
The patent enables flexible parameter changes in the template generation process to adapt to new banknote designs. The system can modify extraction parameters, feature selection criteria, and template processing algorithms to accommodate changes in banknote appearance, allowing rapid adaptation to new denominations and designs without manual reconfiguration by experts.
3Adaptability or versatility
If multiple banknote denominations and orientations are recognized, then versatility is improved, but system complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the banknote recognition task into separate modular components: image acquisition, feature extraction, template matching, and classification. Each component handles a specific aspect of recognition independently, allowing the system to manage multiple denominations and orientations through modular processing rather than a monolithic complex system.
Solution Approach 2:
The system achieves universality by creating a single template-based recognition framework that can handle multiple banknote denominations and orientations. The same automated template generation and matching process works across all banknote types, eliminating the need for separate specialized systems for each denomination while maintaining comprehensive recognition capability.
Data Source
AI summary
A technique for use in automated recognition of a media item involves accessing a template that includes multiple segmentation maps that each is associated with one of multiple classes to which the media item might belong. For each of the multiple classes, the segmentation map is applied to an image of the media item to extract a feature set for the image, the feature set is analyzed, and an assessment is made as to whether the media item belongs to the class.