Banknote Recognition Apparatus Segmentation for Speed
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing paper-sheet recognition apparatuses face challenges in processing speed and total processing amount due to the need to evaluate the entire surface of banknotes and repeat recognition processes for multiple templates, leading to inefficiencies.
Innovation Solution
The apparatus narrows down candidate types based on image data, performing type determination and authenticity recognition concurrently, using partial images and unique codes, and handling data in blocks to reduce processing time and amount.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the entire surface of the banknote is evaluated with a template for recognition, then the recognition accuracy is improved, but the processing time increases
Solution Approach 1:
The banknote evaluation process is segmented into two stages: first evaluating a specific region (such as a corner or edge portion) to narrow down candidate types, and then performing detailed recognition only on the identified type. This segmentation reduces the overall processing time while maintaining recognition accuracy by avoiding unnecessary evaluation of the entire surface for all possible types.
Solution Approach 2:
Instead of performing complete template matching on the entire banknote surface for all candidate types, the system performs partial evaluation on a specific region to narrow down candidates first. This partial action approach reduces processing time while still achieving accurate recognition through subsequent focused evaluation on the narrowed-down candidates.
2Measurement precision
If recognition process is repeated for each of a plurality of templates to identify denomination, then the recognition accuracy is improved, but the total processing amount increases
Solution Approach 1:
The recognition process is segmented into a two-stage approach: first narrowing down candidate types by evaluating specific regions, and then performing detailed template matching only for the narrowed-down candidates. This segmentation reduces the total number of template comparisons needed, thereby improving processing speed while maintaining recognition accuracy.
Solution Approach 2:
The system performs preliminary evaluation of specific regions to narrow down candidate types before conducting the full recognition process. This preliminary action reduces the number of templates that need to be fully processed, improving processing speed without sacrificing the accuracy that would result from complete template matching of all candidates.
3Device complexity
If a narrow aperture sensor is used to sample the banknote, then the device complexity is reduced, but the processing time increases
Solution Approach 1:
The sampling process is segmented into evaluating specific regions of the banknote rather than requiring complete sampling of the entire surface. By focusing on key regions to narrow down candidate types first, the system reduces the effective sampling time needed while maintaining recognition accuracy, thereby reducing the time penalty associated with using a narrow aperture sensor.
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
A paper-sheet recognition apparatus includes a paper-sheet information acquisition unit that acquires paper-sheet information including an image data of the paper sheet; a candidate narrowing-down unit that narrows down a number of type candidates of the paper sheet to a small number of types based on the image data included in the paper-sheet information; a type determining unit that determines one type from the type candidates narrowed down by the candidate narrowing-down unit based on the image data included in the paper-sheet information; authenticity recognition unit that recognizes authenticity of the paper sheet as to each type candidate narrowed down by the candidate narrowing-down unit; an execution instructing unit that issues an instruction such that the type determining unit and the authenticity recognition unit are operated concurrently; and a final judgment unit that performs a final judgment on the paper sheet by combining the type determined by the type determining unit and authenticity recognition result corresponding to the type from among authenticity recognition results of the candidate types recognized by the authenticity recognition unit.