Pupil Diameter Signal Processing for Work Performance Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating task performance by measuring pupil diameter variations require additional devices to measure gaze brightness, increasing cost and inconvenience.
Innovation Solution
An estimation device that processes time-series pupil diameter data using different time windows to separate macro and micro variations, estimating task performance based on micro variations without measuring gaze brightness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a line-of-sight position measuring device and a brightness measuring device are used to accurately acquire gaze brightness, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The invention extracts and removes the harmful macro variation component (caused by gaze brightness changes) from the pupil diameter time-series data through signal processing. By applying a first time window to capture macro variations and a second time window to capture micro variations, then subtracting the former from the latter, the system isolates the micro variation that reflects task performance without needing to measure gaze brightness directly.
Solution Approach 2:
The invention introduces time window smoothing as an intermediary processing step between raw pupil diameter measurement and task performance estimation. The first time window (larger) acts as a mediator to capture and represent the macro variation component, while the second time window (smaller) captures the micro variation component. The difference between these two smoothed signals serves as an intermediary representation that eliminates the need for direct brightness measurement.
2Measurement precision
If multiple devices are used to measure both pupil diameter and gaze brightness, then measurement precision is improved, but ease of operation deteriorates
Solution Approach 1:
The invention extracts the harmful macro variation component from the pupil diameter signal through mathematical processing. By applying a first time window to obtain a smoothed signal representing macro variations and subtracting it from the original or second-smoothed signal, the system isolates micro variations that reflect task performance. This extraction eliminates the need for additional brightness measurement devices and simplifies the operational process.
Solution Approach 2:
The invention replaces the mechanical/physical system of multiple measurement devices (line-of-sight tracker and brightness sensor) with a signal processing approach using time window smoothing and subtraction. This substitution transforms a complex multi-device measurement system into a simpler computational process that operates on existing pupil diameter data, significantly improving ease of operation.
3Reliability
If the pupil diameter data is smoothed by a large time window to remove macro variations, then reliability is improved, but measurement precision of micro variations deteriorates
Solution Approach 1:
The invention segments the pupil diameter time-series data processing into two distinct smoothing operations with different time window sizes. The first time window (larger) segments out the macro variation component, while the second time window (smaller) preserves the micro variation component. By computing the difference between these two segmented representations, the system reliably removes temporary variation influence while maintaining micro variation detection precision.
Solution Approach 2:
The invention applies different smoothing characteristics (different time window sizes) to different aspects of the signal analysis. The larger first time window provides strong smoothing for macro variation removal, while the smaller second time window provides lighter smoothing to preserve micro variations. This local differentiation of smoothing quality allows simultaneous achievement of reliable macro variation removal and precise micro variation detection.
Data Source
AI summary
An estimation device includes processing circuitry configured to acquire time-series data of a pupil diameter during work of a subject of work performance estimation, smooth the time-series data of the pupil diameter of the subject by a first time window, smooth the time-series data of the pupil diameter by a second time window that is a time window larger than the first time window, calculate a difference between time-series data of the pupil diameter smoothed by the first time window and time-series data of the pupil diameter smoothed by the second time window, and estimate a level of work performance of the subject based on a magnitude of the difference between the time-series data of the pupil diameter smoothed by the first time window and the time-series data of the pupil diameter smoothed by the second time window, and output a result of the estimation.