Object Registration via Intra-Object Cross-Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing systems for determining authenticity by imaging random patterns on objects require significant man-hours for registration, as they typically involve comparing multiple images from different objects, which is inefficient.
Innovation Solution
An information processing apparatus and method that extracts a first region of predetermined size from an object's image, specifies a second region for comparison, calculates similarity using cross-correlation values, and registers the image if the similarity meets a predetermined standard, reducing the need for multiple object comparisons.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple images from different objects are used for registration verification, then the reliability of the registered image is improved, but the man-hours required for registration increase significantly
Solution Approach 1:
The patent segments the verification process into two distinct parts: (1) intra-object verification by comparing multiple regions from the same object, and (2) inter-object verification by comparing with previously registered objects. This segmentation allows the system to perform automated intra-object verification using cross-correlation, significantly reducing manual intervention while maintaining reliability through the additional inter-object comparison step.
Solution Approach 2:
The system performs self-service by automatically verifying the suitability of a target image for registration using cross-correlation comparison of multiple regions from the same object. This automated self-verification eliminates the need for manual review of each registered image, reducing man-hours while maintaining quality control through the mathematical rigor of cross-correlation analysis.
2Productivity
If a single object's image is registered without comparing multiple objects, then the registration process is simplified and man-hours are reduced, but the reliability of authenticity determination may be compromised
Solution Approach 1:
The patent applies preliminary action by performing automated intra-object verification through cross-correlation comparison before the image is registered. This preliminary verification ensures the image quality and randomness are sufficient, so that when the image is later used for authenticity determination, the reliability is maintained even though only a single object is registered. The preliminary action prevents poor-quality images from being registered in the first place.
Solution Approach 2:
The system uses feedback by calculating cross-correlation values between multiple regions of the target object and using this information to determine whether the image is suitable for registration. This feedback mechanism provides objective, quantitative criteria for registration suitability, ensuring that only images meeting predetermined thresholds are registered, thereby maintaining reliability while improving productivity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An information processing apparatus includes a processor configured to: extract a first region showing a feature of an object to be registered and having a predetermined size from an object surface image obtained by imaging a surface of the object; specify, from the object surface image, a second region to be compared with the first region and having a same size as the first region; derive a similarity between the first region and the second region; and register, as a registered image, an image of the first region in a case where the similarity between the first region and the second region satisfies a predetermined standard.