Object Detection in Thermal Spectrum for Advanced Driver-Assistance Systems (ADAS)

AI-based smart thermal perception systems can cater to the limitations of conventional imaging sensors by providing a more reliable data source in low-lighting conditions and adverse weather conditions. This research evaluates and modifies the state-of-the-art object detection and classifier framewo...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Muhammad Ali Farooq, Peter Corcoran, Cosmin Rotariu, Waseem Shariff
Formato: article
Lenguaje:EN
Publicado: IEEE 2021
Materias:
Acceso en línea:https://doaj.org/article/a938dbce487c442981ef89c9254f67b1
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:a938dbce487c442981ef89c9254f67b1
record_format dspace
spelling oai:doaj.org-article:a938dbce487c442981ef89c9254f67b12021-12-02T00:00:21ZObject Detection in Thermal Spectrum for Advanced Driver-Assistance Systems (ADAS)2169-353610.1109/ACCESS.2021.3129150https://doaj.org/article/a938dbce487c442981ef89c9254f67b12021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9618926/https://doaj.org/toc/2169-3536AI-based smart thermal perception systems can cater to the limitations of conventional imaging sensors by providing a more reliable data source in low-lighting conditions and adverse weather conditions. This research evaluates and modifies the state-of-the-art object detection and classifier framework for thermal vision with seven key object classes in order to provide superior thermal sensing and scene understanding input for advanced driver-assistance systems (ADAS). The networks are trained on public datasets and is validated on test data with three different test approaches which include test-time augmentation, test-time with no augmentation, and test-time with model ensembling. Additionally, a new model ensemble-based inference engine is proposed, and its efficacy is tested on locally gathered novel test data comprising of 20K thermal frames captured with an uncooled LWIR prototype thermal camera in challenging weather and environmental scenarios. The performance analysis of trained models is investigated by computing precision, recall, and mean average precision scores (mAP). Furthermore, the smaller network variant of thermal-YOLO architecture is optimized using TensorRT inference accelerator, which is then deployed on GPU and resource-constrained edge hardware Nvidia Jetson Nano. This is implemented to explicitly reduce the inference time on GPU as well as on Nvidia Jetson Nano to evaluate the feasibility for added real-time onboard installations.Muhammad Ali FarooqPeter CorcoranCosmin RotariuWaseem ShariffIEEEarticleThermal-infraredobject detectionadvanced driver-assistance systemsdeep learningedge computingElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 156465-156481 (2021)
institution DOAJ
collection DOAJ
language EN
topic Thermal-infrared
object detection
advanced driver-assistance systems
deep learning
edge computing
Electrical engineering. Electronics. Nuclear engineering
TK1-9971
spellingShingle Thermal-infrared
object detection
advanced driver-assistance systems
deep learning
edge computing
Electrical engineering. Electronics. Nuclear engineering
TK1-9971
Muhammad Ali Farooq
Peter Corcoran
Cosmin Rotariu
Waseem Shariff
Object Detection in Thermal Spectrum for Advanced Driver-Assistance Systems (ADAS)
description AI-based smart thermal perception systems can cater to the limitations of conventional imaging sensors by providing a more reliable data source in low-lighting conditions and adverse weather conditions. This research evaluates and modifies the state-of-the-art object detection and classifier framework for thermal vision with seven key object classes in order to provide superior thermal sensing and scene understanding input for advanced driver-assistance systems (ADAS). The networks are trained on public datasets and is validated on test data with three different test approaches which include test-time augmentation, test-time with no augmentation, and test-time with model ensembling. Additionally, a new model ensemble-based inference engine is proposed, and its efficacy is tested on locally gathered novel test data comprising of 20K thermal frames captured with an uncooled LWIR prototype thermal camera in challenging weather and environmental scenarios. The performance analysis of trained models is investigated by computing precision, recall, and mean average precision scores (mAP). Furthermore, the smaller network variant of thermal-YOLO architecture is optimized using TensorRT inference accelerator, which is then deployed on GPU and resource-constrained edge hardware Nvidia Jetson Nano. This is implemented to explicitly reduce the inference time on GPU as well as on Nvidia Jetson Nano to evaluate the feasibility for added real-time onboard installations.
format article
author Muhammad Ali Farooq
Peter Corcoran
Cosmin Rotariu
Waseem Shariff
author_facet Muhammad Ali Farooq
Peter Corcoran
Cosmin Rotariu
Waseem Shariff
author_sort Muhammad Ali Farooq
title Object Detection in Thermal Spectrum for Advanced Driver-Assistance Systems (ADAS)
title_short Object Detection in Thermal Spectrum for Advanced Driver-Assistance Systems (ADAS)
title_full Object Detection in Thermal Spectrum for Advanced Driver-Assistance Systems (ADAS)
title_fullStr Object Detection in Thermal Spectrum for Advanced Driver-Assistance Systems (ADAS)
title_full_unstemmed Object Detection in Thermal Spectrum for Advanced Driver-Assistance Systems (ADAS)
title_sort object detection in thermal spectrum for advanced driver-assistance systems (adas)
publisher IEEE
publishDate 2021
url https://doaj.org/article/a938dbce487c442981ef89c9254f67b1
work_keys_str_mv AT muhammadalifarooq objectdetectioninthermalspectrumforadvanceddriverassistancesystemsadas
AT petercorcoran objectdetectioninthermalspectrumforadvanceddriverassistancesystemsadas
AT cosminrotariu objectdetectioninthermalspectrumforadvanceddriverassistancesystemsadas
AT waseemshariff objectdetectioninthermalspectrumforadvanceddriverassistancesystemsadas
_version_ 1718404006651363328