Gate-Specific Face Authentication Collation System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In walkthrough face authentication systems with multiple gates, there is a risk of erroneous recognition of individuals due to overlapping images, and the processing load on information processing apparatuses is consistently high as all individuals in the image are subjected to collation, regardless of their proximity to the gate.
Innovation Solution
A collation system that uses imaging means to capture images of pre-passage areas for each gate, with a collation means initiating the collation process based on conditions set for each area of the captured image, determining the propriety of passage only for targets within specific areas, thereby reducing unnecessary collation processes and processing load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If collation is performed on all persons shown in the input image, then authentication accuracy is maintained, but processing load on the information processing apparatus increases
Solution Approach 1:
The patent divides the captured image into multiple regions corresponding to different gates, and performs collation only on persons located in the region associated with the gate being processed. This segmentation approach reduces the number of persons subjected to collation while maintaining authentication accuracy for the relevant gate.
Solution Approach 2:
The patent applies different collation strategies to different regions of the image based on their spatial relationship with the gate. Persons in the foreground region near the gate are subject to collation, while persons in the background or adjacent lanes are excluded, creating local quality differentiation in the processing approach.
2Reliability
If collation is performed on all persons in the image, then no persons are missed, but erroneous recognition of persons moving toward adjacent gates occurs
Solution Approach 1:
The patent segments the image into distinct regions for different gates and uses this spatial information to determine which persons should be collated. By restricting collation to persons in the appropriate region, the system avoids erroneous recognition of persons moving toward adjacent gates while maintaining accurate identification of persons approaching the current gate.
Solution Approach 2:
The patent introduces an intermediary mechanism that uses image region information as a filter or mediator between the captured image and the collation process. This intermediary step determines eligibility for collation based on spatial position, preventing erroneous matching while ensuring valid persons are not overlooked.
Data Source
AI summary
A collation system of the present invention includes an imaging means for acquiring a captured image of a pre-passage side area with respect to a gate, a collation means for performing a collation process between a previously registered target and a target in the captured image, and a determination means for determining propriety of passage of the target with respect to the gate, on the basis of a result of the collation process. The collation means initiates the collation process on the basis of a condition, set to each area of the captured image, for the target located in the area.