All-Weather Object Recognition Using Radar and Infrared Sensing

TL;DR

This paper presents a novel approach combining radar and infrared sensing with deep learning to improve object recognition in autonomous vehicles under adverse weather conditions, outperforming traditional optical sensors.

Contribution

It introduces new sensing methodologies using polarised IR and radar data, along with a large-scale dataset for robust object detection in various weather scenarios.

Findings

Radar and IR sensors outperform optical sensors in bad weather

Deep neural networks effectively recognize objects using polarised IR and radar data

A new large-scale dataset demonstrates robustness of radar in adverse conditions

Abstract

Autonomous cars are an emergent technology which has the capacity to change human lives. The current sensor systems which are most capable of perception are based on optical sensors. For example, deep neural networks show outstanding results in recognising objects when used to process data from cameras and Light Detection And Ranging (LiDAR) sensors. However these sensors perform poorly under adverse weather conditions such as rain, fog, and snow due to the sensor wavelengths. This thesis explores new sensing developments based on long wave polarised infrared (IR) imagery and imaging radar to recognise objects. First, we developed a methodology based on Stokes parameters using polarised infrared data to recognise vehicles using deep neural networks. Second, we explored the potential of using only the power spectrum captured by low-THz radar sensors to perform object recognition in a controlled scenario. This latter work is based on a data-driven approach together with the development of a data augmentation method based on attenuation, range and speckle noise. Last, we created a new large-scale dataset in the "wild" with many different weather scenarios (sunny, overcast, night, fog, rain and snow) showing radar robustness to detect vehicles in adverse weather. High resolution radar and polarised IR imagery, combined with a deep learning approach, are shown as a potential alternative to current automotive sensing systems based on visible spectrum optical technology as they are more robust in severe weather and adverse light conditions.