Presentation
13 June 2022 Multimodal realtime pedestrian detection using radar-camera fusion with clustering
Author Affiliations +
Abstract
Autonomous vehicles design and development can move safely on roads while sensing the environment to focus on pedestrian detection systems so that people can be detected as quickly and accurately as possible. First, however, it is critical to examine the pedestrians themselves and their color, which benefits from being insensitive to changes in scale and partial occlusion. Moreover, human skin detection has proven to be a tough challenge since skin color can vary considerably in appearance due to various factors such as lighting, race, and imaging circumstances. Unfortunately, human skin detection has not been thoroughly investigated in this circumstance, and it appears that many studies do not address this systematically when it comes to pedestrian detection systems for autonomous cars. To overcome this issue, we are using a Radar-Camera fusion technique to predict obstacles in various daylight situations.
Conference Presentation
© (2022) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Kirsnaragavan Arudpiragasam Kannuri, Taraka Rama Krishna Kanth, Klaus Schwarz, Michael Hartmann, and Reiner M. Creutzburg "Multimodal realtime pedestrian detection using radar-camera fusion with clustering", Proc. SPIE PC12100, Multimodal Image Exploitation and Learning 2022, PC1210005 (13 June 2022); https://doi.org/10.1117/12.2619013
Advertisement
Advertisement
KEYWORDS
Skin

Unmanned vehicles

Convolutional neural networks

Environmental sensing

Light sources and illumination

Roads

Sensing systems

Back to Top