Disorderly Biometric Boarding Through Trajectory Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric boarding systems face inefficiencies due to the 'stacking problem' where passengers stand too close, causing identification confusion and delays, requiring strict spacing and manual intervention to maintain order, which increases processing time and reliance on staff.
Innovation Solution
A method using two-dimensional and three-dimensional image data to track passengers, detect discontinuities in their trajectories, and assign unique identifiers to resolve identity confusion, allowing disorderly boarding without strict spacing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If strict spacing between passengers is maintained to avoid identification confusion, then identification accuracy is improved, but boarding time increases significantly
Solution Approach 1:
The patent transitions from 2D face recognition to 3D depth-aware recognition. By incorporating depth information from 3D cameras or time-of-flight sensors, the system can distinguish between faces at different distances from the camera, allowing passengers to stand closer together without causing identification confusion. This dimensional addition resolves the contradiction by maintaining accuracy while reducing spacing requirements.
Solution Approach 2:
The system performs preliminary depth assessment and trajectory prediction before final identification. By pre-evaluating the spatial position and movement trajectory of each passenger, the system can anticipate potential stacking issues and adjust processing accordingly, preventing identification errors before they occur and maintaining fast processing speeds.
2Reliability
If passengers are required to stand further apart to avoid stacking problems, then identification reliability is improved, but process efficiency deteriorates
Solution Approach 1:
The patent replaces the mechanical approach of physically spacing passengers apart with an optical/information-based solution using 3D depth sensing and trajectory analysis. This substitution maintains identification reliability through accurate depth-based discrimination while eliminating the need for reduced passenger density, thereby preserving process efficiency.
Solution Approach 2:
The system introduces depth information and trajectory data as intermediary elements between the camera and the identification process. These intermediaries provide additional contextual information that enhances identification reliability without requiring physical spacing, thus maintaining high process efficiency.
3Measurement precision
If biometric cameras are placed at a right angle to the queue to reduce stacking problems, then identification accuracy is improved, but passenger alignment difficulty increases
Solution Approach 1:
Instead of changing the camera's angular position relative to the queue, the patent adds the depth dimension to the existing forward-facing camera setup. This allows the camera to maintain its natural alignment with passenger movement while using depth information to distinguish between closely spaced faces, thereby maintaining accuracy without increasing alignment difficulty.
4Reliability
If manual intervention is increased to maintain queue spacing and prevent stacking, then identification reliability is improved, but automation level decreases
Solution Approach 1:
The system enables self-service by using automated 3D depth sensing and trajectory analysis to prevent stacking problems without requiring manual intervention. The technology autonomously distinguishes between passengers based on their spatial positions and movement patterns, maintaining identification reliability while maximizing automation and minimizing staff involvement.
Solution Approach 2:
The system continuously monitors passenger positions and trajectories in real-time, providing feedback that enables dynamic adjustment of identification processing. This automated feedback loop maintains identification reliability by detecting and resolving potential stacking issues without requiring manual intervention, thereby preserving high automation levels.
Data Source
AI summary
Disclosed is a method for controlling access for at least one tracked object, including: acquiring or receiving a series of two-dimensional images assumed to be taken of the at least one tracked object, and also position data in respect of the at least one tracked object; assigning a unique tracking identifier to the at least one tracked object; determining a trajectory of the at least one tracked object from the position data; determining if there is a discontinuity in the trajectory or data computed from the trajectory, and if a discontinuity is detected, acquiring or receiving one or more new images of the at least tracked object, and assigning a new unique tracking identifier to said at least one tracked object; and determining whether access should be allowed, on the basis of at least one of the one or more new images if discontinuity is detected, or on the basis of the at least one image from the series of two-dimensional images if discontinuity is not detected.