Paper Sheet Recognition Adaptive Block Division
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing paper sheet recognition methods face delays and errors when dealing with numerous types of paper sheets, particularly banknotes, due to repetitive processing and potential misjudgment of modified or partial banknotes.
Innovation Solution
A paper sheet recognition apparatus that selects candidate types based on density and direction features of uniformly divided image blocks, and adaptively divides the input image according to reference image features for detailed matching, using multiple image types (infrared and visible reflection/transmission) for efficient narrowing down.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the narrowing-down process repeats for each combination of two judgment candidates, then the recognition accuracy is improved, but the processing time increases significantly when the number of judgment candidates is large
Solution Approach 1:
The patent segments the recognition process into two distinct stages: a first judgment process that performs preliminary filtering to eliminate clearly non-matching candidates, and a second judgment process that performs detailed comparison only on the remaining candidates. This segmentation reduces the number of repetitive comparisons needed while maintaining overall recognition accuracy.
Solution Approach 2:
The first judgment process acts as a preliminary action that pre-filters the candidate banknotes before the second judgment process. By performing initial elimination based on less computationally intensive features, the system reduces the workload for the subsequent detailed judgment, thereby decreasing total processing time without sacrificing accuracy.
2Productivity
If preprocessing specifies one judgment candidate from multiple candidates, then the processing speed is improved, but the recognition accuracy deteriorates due to potential judgment errors
Solution Approach 1:
The patent divides the candidate selection process into two stages: first judgment that performs preliminary filtering to narrow down candidates, and second judgment that performs detailed verification. This segmentation allows the system to benefit from the speed of preliminary filtering while ensuring accuracy through subsequent detailed comparison on the reduced candidate set.
Solution Approach 2:
The first judgment process serves as an intermediary that bridges the gap between rapid preprocessing and accurate detailed judgment. It performs initial filtering to create a reduced candidate set that is then subjected to thorough verification, ensuring that no potentially correct candidates are incorrectly eliminated while still achieving faster processing than direct detailed comparison of all candidates.
3Measurement precision
If multiple types of images are used for recognition, then the recognition accuracy is improved, but the processing complexity increases
Solution Approach 1:
The patent segments the image processing into two distinct judgment stages: first judgment that processes multiple image types for preliminary candidate elimination, and second judgment that performs detailed comparison. This segmentation manages complexity by handling different levels of analysis separately, allowing comprehensive multi-image analysis without overwhelming computational burden.
Solution Approach 2:
The first judgment process performs preliminary analysis of multiple image types to eliminate clearly non-matching candidates before the second judgment process. This preliminary action reduces the number of detailed comparisons needed across multiple image types, thereby managing processing complexity while still leveraging the accuracy benefits of multi-image analysis.
Data Source
AI summary
A paper sheet recognition apparatus recognizes a type of a paper sheet in an input image by matching the input image of the paper sheet with reference images of a plurality of paper sheets. The apparatus includes a candidate selecting unit that selects a predetermined number of candidate types of the paper sheet based on a density feature and direction features of each block acquired by uniformly dividing the input image and the reference images; and a detailed judgment unit that adaptively divides the input image into blocks according to features of a reference image corresponding to each type of paper sheets selected by the candidate selecting unit, and performs a detailed judgment based on matching values between corresponding blocks of the divided input image and each of the reference images.