Drowsiness Estimation Using Blink-Filtered Eye Openness Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating drowsiness from eye-openness width signals suffer from inaccuracies due to variations in time window width and sampling rate, making them unreliable across different conditions.
Innovation Solution
A drowsiness estimating device and method that filters eye-openness width signals to remove blinking artifacts, calculates features such as variation within a specified window, and uses these features to estimate drowsiness levels, independent of time window width and sampling rate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing methods use fixed time window width and sampling rate for drowsiness estimation, then the estimation process is simple, but the accuracy decreases under varying conditions
Solution Approach 1:
The patent applies dynamics by making the time window width and sampling rate adjustable rather than fixed. The system dynamically adapts these parameters based on the specific application scenario and signal characteristics, allowing optimal drowsiness estimation accuracy across different conditions without being constrained by predetermined fixed values
Solution Approach 2:
The patent implements parameter changes by systematically varying time window width and sampling rate to identify optimal values for different scenarios. The method explores parameter space to find combinations that maximize estimation accuracy, transforming static parameters into optimizable variables that can be tuned for specific applications
2Measurement precision
If high sampling rate and long time window are used to improve drowsiness estimation accuracy, then measurement precision increases, but device complexity and processing load increase
Solution Approach 1:
The patent applies this principle by using computationally efficient algorithms and optimized processing approaches that achieve accurate drowsiness estimation without requiring excessive computational resources. The method prioritizes cost-effective solutions that balance accuracy with processing requirements, avoiding unnecessarily complex hardware or algorithms
Solution Approach 2:
The patent implements partial action by selecting sampling rates and time window lengths that are sufficient for accurate drowsiness detection without being excessively high. The method identifies the minimum necessary processing parameters needed to achieve reliable results, avoiding the overhead of overly aggressive sampling or unnecessarily long analysis windows
Data Source
AI summary
This drowsiness estimating device estimates a subject's drowsiness from a time-series signal of the subject's eye-openness width. A filtering circuitry filters the time-series signal of the eye-openness width to eliminate signal changes due to the subject blanking and outputs the filtered time-series signal of eye-openness. A feature calculator calculates a feature from at least the filtered time-series signal of the eye-openness width. A drowsiness estimator estimates a drowsiness evaluated value from the feature and outputs an estimated result. The feature calculator includes at least a first feature calculation circuit that calculates a variation of the filtered time-series signal of the eye-openness width within a feature calculation window width and outputs the variation as a first feature.