Continuous Authentication Camera Control for Multi-Person Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In continuous authentication systems, the detection of multiple individuals in an image captured by a camera during check-in can lead to inaccurate identification of the target person, as biometric data of non-target individuals may be erroneously acquired, reducing the accuracy of continuous authentication.
Innovation Solution
A biometric authentication system that includes a biometric sensor, a linking camera, and a tracking camera, with an information processing device controlling the detection area to ensure the target person is accurately identified by determining their direction and number within the image, thereby extracting accurate feature information for continuous authentication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If the camera captures a wide area to detect multiple people, then the coverage area is improved, but the detection precision of the target person deteriorates due to interference from other people
Solution Approach 1:
The patent divides the detection area into multiple regions and identifies the target person by their spatial relationship with the authentication device. When multiple people are detected, the system segments the detection results and selects the person located at the authentication device side, thereby maintaining detection precision while preserving wide coverage capability.
Solution Approach 2:
The patent applies different detection criteria to different regions of the image. The target detection area is specifically defined as the region where the authentication device is located, and the system prioritizes detecting the person in this specific local area over other regions, thus improving detection precision for the target person while maintaining overall coverage.
2Measurement precision
If the detection area is narrowed to focus on the target person, then the detection precision is improved, but the coverage area is reduced
Solution Approach 1:
The patent performs preliminary detection of multiple people in the wide area, then uses this information to determine the target person's location relative to the authentication device. This preliminary wide-area detection enables subsequent precise targeting without sacrificing coverage, as the system uses the preliminary results to guide the final selection.
Solution Approach 2:
The patent adds a spatial dimension to the detection criteria by considering the position of detected people relative to the authentication device. Instead of merely detecting people in a narrow area, the system evaluates the spatial coordinates and selects the person in the specific region near the authentication device, thereby achieving precise target detection while maintaining wide coverage through multi-dimensional selection criteria.
3Quantity of substance
If biometric data is collected from all detected people, then the quantity of data is improved, but the reliability of authentication deteriorates due to inclusion of non-target individuals
Solution Approach 1:
The patent extracts and selects only the biometric data of the target person from the set of all detected people. By using spatial information and the authentication device location as criteria, the system extracts the correct person's data while excluding non-target individuals, thus maintaining high reliability while collecting sufficient biometric data for authentication.
Solution Approach 2:
The patent uses the authentication device location and spatial relationships as an intermediary criterion to filter and select the target person's biometric data. This intermediary spatial information acts as a mediator between the raw detection data of all people and the final selected biometric data, ensuring reliability by using the spatial mediator to identify the correct target person.
Data Source
AI summary
In a control method, when a plurality of people are detected from an image including a person to be authenticated who has been successfully authenticated using identification information, a target detection area for detecting the person in the image is controlled.