Preventing Handheld Phone Distraction for Drivers by Sensing the Gripping Hand
URL: http://arxiv.org/abs/2111.05738v1
archive: archived pipeline: cataloged verified
Abstract
Handheld phone distraction is the leading cause of traffic accidents. However, few efforts have been devoted to detecting when the phone distraction happens, which is a critical input for taking immediate safety measures. This work proposes a phone-use monitoring system, which detects the start of the driver's handheld phone use and eliminates the distraction at once. Specifically, the proposed system emits periodic ultrasonic pulses to sense if the phone is being held in hand or placed on support surfaces (e.g., seat and cup holder) by capturing the unique signal interference resulted from the contact object's damping, reflection and refraction. We derive the short-time Fourier transform from the microphone data to describe such impacts and develop a CNN-based binary classifier to discriminate the phone use between the handheld and the handsfree status. Additionally, we design an adaptive window-based filter to correct the classification errors and identify each handheld phone distraction instance, including its start, end, and duration. Extensive experiments with fourteen people, three phones and two car models show that our system achieves 99% accuracy of recognizing handheld phone-use instances and 0.76-second median error to estimate the distraction's start time.
Summary
Wang, Huang & Wang (2021, arXiv 2111.05738, LSU) propose a phone-distraction detection system that emits periodic ultrasonic pulses from a smartphone speaker and uses microphone reception to capture damping/reflection/refraction signatures specific to a gripping hand versus support surfaces (seat, cup holder, mount, pocket). Short-time Fourier transforms feed a CNN binary classifier (handheld vs handsfree), and an adaptive window-based filter corrects classification errors and segments each distraction instance (start, end, duration). Evaluation with 14 participants, three phones, and two car models reported 99% accuracy in recognizing handheld phone-use instances and 0.76-second median error on detecting distraction onset.
Key finding
Ultrasonic acoustic sensing on a stock smartphone can identify when the driver is gripping the phone (vs placing it on a surface) with 99% accuracy and sub-second timing precision, enabling real-time intervention against handheld-phone distraction without dedicated hardware.
Methodology
lab_experiment
Sample size: 14
Quality score: 5 / 5