Verification Device Normalizing Input Patterns Using Multiple Reference Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing pattern verification systems face challenges in accurately determining similarities between input patterns due to variations in photographic conditions, leading to erroneous determinations, as they rely solely on similarity calculations between input patterns without adequate normalization or correct comparative pattern generation.
Innovation Solution
A verification device that calculates similarities between input patterns and a set of reference information, using a plurality of reference patterns stored under specific conditions to normalize and correct for variations in photographic conditions, thereby reducing erroneous determinations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If similarity calculation is performed directly between input patterns without normalization, then verification process is simple and fast, but verification accuracy deteriorates due to photographic condition variations
Solution Approach 1:
The patent introduces reference patterns as an intermediary element between input patterns. Instead of directly comparing input patterns, the system calculates similarities between each input pattern and multiple reference patterns, then integrates these similarities. This intermediary approach normalizes the comparison process and eliminates direct dependency on photographic conditions, thereby improving verification accuracy while maintaining a manageable process through systematic integration.
Solution Approach 2:
The patent transforms the verification approach by changing the parameter of comparison from direct input-pattern-to-input-pattern similarity to input-pattern-to-reference-pattern similarity integration. By calculating similarities with multiple reference patterns under controlled conditions and integrating these results, the system normalizes the verification process against photographic condition variations, improving reliability without excessive complexity.
2Measurement precision
If multiple reference patterns are used for similarity calculation, then verification accuracy improves by normalizing photographic condition variations, but processing time and computational cost increase
Solution Approach 1:
The patent applies preliminary action by pre-storing multiple reference patterns under controlled photographic conditions before verification. These reference patterns are prepared in advance and stored for future use. During verification, the system simply retrieves and compares against these pre-prepared references, avoiding the need to generate reference patterns in real-time, thus reducing processing time while maintaining high precision through normalized comparisons.
Solution Approach 2:
The system uses a plurality of reference patterns (excessive action) to ensure accurate normalization, but the integration process is designed to be efficient. By calculating similarities with multiple references and integrating them systematically, the patent achieves high measurement precision while the efficient integration method prevents excessive processing time, balancing accuracy and speed.
3Reliability
If similarity threshold is set strictly for accurate verification, then false acceptance rate decreases, but false rejection rate increases leading to more erroneous determinations
Solution Approach 1:
The patent implements feedback through the integration of multiple similarity calculations. Instead of relying on a single threshold comparison, the system calculates similarities with multiple reference patterns and integrates these results to determine final verification outcomes. This feedback mechanism provides a more robust basis for decision-making, reducing both false acceptance and false rejection rates by considering multiple evidence points rather than a single threshold comparison.
Data Source
AI summary
Disclosed is a verification device and the like that suppress an erroneous determination upon determining a difference between input patterns based on a similarity to a reference pattern recorded under a specific condition. A verification device 100 includes a similarity calculation unit 6 calculating similarities S 7 between a set of input information x 110 and y 111 indicating features related to input patterns that are objects of verification and a plurality of types of reference information 112 indicating features related to a reference pattern to be a reference of the verification by using the set of input information x 110 and y 111 and the plurality of types of reference information 112. The calculated similarities S7 are presented to an external device or a user.