EBSnoR: Event-Based Snow Removal by Optimal Dwell Time Thresholding

TL;DR

EBSnoR is an innovative event-based snow removal method that uses dwell time thresholding and hypothesis testing to distinguish snowflakes from background, improving vehicle detection in snowy conditions.

Contribution

The paper introduces a novel snow removal algorithm using dwell time analysis and hypothesis testing on event-based camera data, validated on a new dataset.

Findings

EBSnoR accurately identifies snowflake events.

Preprocessing with EBSnoR enhances car detection accuracy.

The method effectively separates snow from background in event data.

Abstract

We propose an Event-Based Snow Removal algorithm called EBSnoR. We developed a technique to measure the dwell time of snowflakes on a pixel using event-based camera data, which is used to carry out a Neyman-Pearson hypothesis test to partition event stream into snowflake and background events. The effectiveness of the proposed EBSnoR was verified on a new dataset called UDayton22EBSnow, comprised of front-facing event-based camera in a car driving through snow with manually annotated bounding boxes around surrounding vehicles. Qualitatively, EBSnoR correctly identifies events corresponding to snowflakes; and quantitatively, EBSnoR-preprocessed event data improved the performance of event-based car detection algorithms.