Puscul, Dzenana (Graz University of Technology), Schoenlieb, Armin (Infineon), Ramsauer, Andreas (AVL List GmbH), Lex, Cornelia (Graz University of Technology)

On-Road Water Detection and Water Film Height Estimation Using a Time-Of-Flight Camera

Scheduled for presentation during the Poster Session "Perception - Road and weather conditions" (WeBT16), Wednesday, September 25, 2024, 14:30−16:30, Foyer

2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), September 24- 27, 2024, Edmonton, Canada

This information is tentative and subject to change. Compiled on December 26, 2024

Abstract

In this article, a methodology to detect the presence of water on the road with a time-of-flight (ToF) camera is presented and validated with measurements from a proving ground. The water detection algorithm distinguishes between dry and wet roads and is based on the measurement data of two time-of-flight imagers. Furthermore, the water film height is estimated with a feedforward neural network. The presented measurements were carried out on the stationary vehicle. The variations of measured conditions include different lighting conditions (daylight, artificial light at night, dawn period) as well as dry and wet asphalt with different water film levels. The results show that the water detection is reliable and the water film height can be well estimated by the neural network.