Vehicle Detection under Adverse Weather from Roadside LiDAR Data

Jianqing Wu,Rui Yue,Rendong Pi,Yuan Tian,Hao Xu

doi:10.3390/s20123433

Abstract

Roadside light detection and ranging (LiDAR) is an emerging traffic data collection device and has recently been deployed in different transportation areas. The current data processing algorithms for roadside LiDAR are usually developed assuming normal weather conditions. Adverse weather conditions, such as windy and snowy conditions, could be challenges for data processing. This paper examines the performance of the state-of-the-art data processing algorithms developed for roadside LiDAR under adverse weather and then composed an improved background filtering and object clustering method in order to process the roadside LiDAR data, which was proven to perform better under windy and snowy weather. The testing results showed that the accuracy of the background filtering and point clustering was greatly improved compared to the state-of-the-art methods. With this new approach, vehicles can be identified with relatively high accuracy under windy and snowy weather.

Highlights

Adverse weather can negatively influence transportation performance in two aspects: decreasing the operational efficiency and increasing the crash risk
The roadside Light detection and ranging (LiDAR) sensor is able to scan the surfaces of all road vehicles within the detection range by generating 3D point clouds, which provides a perfect solution for filling the data gap of the transition
The results showed that all the LiDAR systems decreased in fog and that changing the internal parameters in the LiDAR could improve their functions under adverse weather

Summary

Introduction

Adverse weather can negatively influence transportation performance in two aspects: decreasing the operational efficiency and increasing the crash risk. The roadside LiDAR sensor is able to scan the surfaces of all road vehicles (including both connected vehicles and unconnected vehicles) within the detection range by generating 3D point clouds, which provides a perfect solution for filling the data gap of the transition. The original method for filtering the background was to search the frames without road users within the detection range [12,13]. Lv et al [17] developed a raster-based (RA) method using the change in point density as a feature for background filtering. It is still necessary to quantitatively analyze the influence of different adverse conditions on the roadside LiDAR and to develop new methods that can accommodate background filtering and point clustering for adverse weather conditions

Background

Point Clustering

Findings

Conclusions and Discussion