Facial Recognition Age Thresholding for Infant Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous facial recognition systems face high error rates when identifying infants and small children due to the similarity in their facial features, leading to misidentification of siblings or different infants.
Innovation Solution
The system determines an estimated biological age of a subject in a current image and calculates a temporal threshold based on this age, comparing it with the age of historical images to accurately associate identifiers, thereby reducing misidentification by considering the rate of physiological change in facial structure over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional facial recognition algorithms are used to identify infants and small children, then the system can process images quickly, but the accuracy of identification deteriorates due to similarity in facial features
Solution Approach 1:
The system dynamically adjusts the temporal threshold based on the estimated biological age of the subject. For infants and young children whose facial features change rapidly, a shorter temporal threshold is applied. For older individuals with stable facial features, a longer temporal threshold is used. This dynamic adaptation resolves the contradiction by making the identification accuracy comparable across different age groups despite varying rates of facial feature change.
Solution Approach 2:
The system changes the parameter of temporal threshold based on the biological age of the subject. By estimating the biological age from the image and comparing it with historical images, the system adjusts the temporal window for identification. This parameter change allows the system to account for rapid facial feature changes in infants while maintaining efficiency for older subjects.
2Measurement precision
If the temporal threshold is extended to accommodate facial feature changes in children, then identification accuracy improves, but the system complexity increases due to age estimation requirements
Solution Approach 1:
The system performs preliminary estimation of biological age from the current image before conducting the identification process. This preliminary action provides the basis for selecting an appropriate temporal threshold, streamlining the overall process despite the added complexity of age estimation.
Solution Approach 2:
The system introduces biological age estimation as an intermediary step between image acquisition and identification. This intermediary provides crucial information about the subject's developmental stage, enabling the system to select an appropriate temporal threshold without requiring complex manual configuration or multiple separate systems.
3Reliability
If biological age estimation is incorporated into the facial recognition system, then misidentification of children is reduced, but the processing time increases
Solution Approach 1:
The system applies partial action by selectively using biological age estimation and temporal threshold adjustment only when needed - specifically when identifying infants and young children. For older individuals with stable facial features, the system can use standard identification processes, thus avoiding unnecessary processing time while maintaining high reliability for the vulnerable age group.
Data Source
AI summary
A facial recognition system determines an estimated biological age of a subject in a current image. A facial recognition controller identifies the subject in the current image using a number of historical images, at least one of which includes the subject and a logically associated identifier. The system compares the estimated biological age of the subject to a defined threshold value. The system determines a temporal difference between a time/date logically associated with the current image and a time/date logically associated with the historical image. If the estimated biological age of the subject in the current image is less than a defined age threshold and the temporal difference between the time/date logically associated with the current image and a time/date logically associated with the historical image is less than a temporal threshold, the identifier logically associated with the historical image is logically associated with the current image.