Abstract
Deviation analysis describes the variation and offset between a target model and a reference model, which is essential for assessing the fit of wearable products. Current general-purpose CAD software often produces discontinuous data when analyzing the complex interface between earphones and human ears, leading to inaccuracies. This study proposes a novel deviation analysis algorithm specifically designed for earphone fit. The algorithm accurately classifies interference and gap sets, ensuring precise deviation calculations. Validation through human trials and comparison with standard software results demonstrates that this new approach significantly improves the accuracy and usability of virtual fit assessments.
Introduction
With the rise of True Wireless Stereo (TWS) earphones, ergonomic fit has become a critical factor in consumer satisfaction. Designers rely on virtual deviation analysis—measuring the distance between 3D scans of ears and earphones—to predict comfort and stability. However, standard CAD tools often struggle with the "double-counting" of points when surfaces interpenetrate. This results in data "jumps" and fragmented visual maps that do not reflect physical reality, making it difficult for designers to identify actual pressure points or loose gaps.
Methodology
The proposed algorithm addresses these shortcomings through a specialized five-step process designed for complex ear geometry.
- Point Cloud Preprocessing: The ear model serves as the reference, and the earphone as the measurement model. Redundant data is removed to streamline calculation.
- Distance Vector Generation: For each point on the reference model, the algorithm calculates a vector to the nearest triangular facet on the measurement model.
- Interference vs. Gap Classification: This core step determines whether a point represents "interference" (pressure) or a "gap" (space). By analyzing the angle between the distance vector and the surface normal, the algorithm distinguishes between overlapping and non-overlapping surfaces.
- Validity Filtering: To ensure "one point, one calculation," the algorithm filters out physically impossible vectors, such as those passing through the model to distant surfaces.
- Statistical Visualization: Results are mapped onto the 3D model as a continuous color-coded heat map.
Experimental Validation
The algorithm was tested using 3D scans from 34 participants and the Apple AirPods 2. Subjective comfort and stability were measured using a modified Borg scale (0–10) across eight ear regions.
Key findings include:
- Improved Correlation: Unlike standard CAD tools, which showed no significant correlation with user experience, the new algorithm's metrics (e.g., Maximum Gap and Average Value) showed a significant moderate correlation ($r_s \approx -0.4$) with subjective fit scores.
- Stability Prediction: The "Gap Standard Deviation" metric proved to be a reliable predictor of perceived stability ($r_s = 0.482$).
- Regional Insights: The analysis highlighted that fit in specific areas, such as the incisura intertragica, is a primary driver of overall comfort.
Conclusion
The proposed algorithm provides a more accurate and physically meaningful way to analyze earphone fit compared to general-purpose CAD tools. By eliminating data discontinuities and aligning with subjective user feedback, it offers a robust tool for ergonomic optimization. Future work will aim to incorporate material properties (soft tissue deformation) and non-geometric factors like friction and weight into the model.
Reference: YAN Yan, LIU Yonghong, WANG Haining. An Accurate Deviation Analysis Algorithm for Earphone Fit Analysis. Journal of Mechanical Engineering, 2023, 59(11): 54-64.
Summary updated