University of Twente Student Theses

Login
As of Friday, 8 August 2025, the current Student Theses repository is no longer available for thesis uploads. A new Student Theses repository will be available starting Friday, 15 August 2025.

Earable-Based Visual Distraction Monitoring in Cyclists

Balakrishnan, Sidhharth (2025) Earable-Based Visual Distraction Monitoring in Cyclists.

[img] PDF
2MB
Abstract:Visual distractions among cyclists significantly lower cyclists’ situational awareness, heightening the risk of accidents. This paper proposes the utilization of an open-source OpenEarable device, which is equipped with onboard inertial measurement units (IMU), as an easy and non-invasive method for detecting visual distractions through the quantification of head movements that are indicative of behaviours associated with visual distraction. Head movement patterns of 20 subjects were recorded using earable IMU sensors in naturalistic cycling scenarios. Classical machine learning and deep learning models were employed to analyze the collected data and identify patterns characteristic of visual distractions in cyclists. Among machine learning models, Support Vector Machine (SVM) achieved the highest F1-score (85%) with a fair Kappa score (0.59). In deep learning models, Convolutional Neural Network (CNN) offers the best F1-score (87%) and a substantial Kappa score (0.74). To asses edge-deployment feasibility, the models are further optimized and evaluated for deployment on Raspberry Pi. The LinearSVC variant of the SVM model offers the best tradeoff between model size, inference time, and classification performance. Whereas, for CNN, quantization techniques like Post Training Quantization (PTQ) and Quantization-Aware Training (QAT) reduce the model size without sacrificing performance. These results highlight the potential of earable devices for realtime distraction detection and provide a foundation for future wearable mobility safety systems.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Embedded Systems MSc (60331)
Link to this item:https://purl.utwente.nl/essays/106412
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page