Using layer-wise training for Road Semantic Segmentation in Autonomous Cars

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Autoren

  • Shahrzad Shashaani
  • Mohammad Teshnehlab
  • Amirreza Khodadadian
  • Maryam Parvizi
  • Thomas Wick
  • Nima Noii

Externe Organisationen

  • K.N. Toosi University of Technology
Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Seiten (von - bis)46320 - 46329
Seitenumfang10
FachzeitschriftIEEE ACCESS
Jahrgang11
PublikationsstatusVeröffentlicht - 10 März 2023

Abstract

A recently developed application of computer vision is pathfinding in self-driving cars. Semantic scene understanding and semantic segmentation, as subfields of computer vision, are widely used in autonomous driving. Semantic segmentation for pathfinding uses deep learning methods and various large sample datasets to train a proper model. Due to the importance of this task, accurate and robust models should be trained to perform properly in different lighting and weather conditions and in the presence of noisy input data. In this paper, we propose a novel learning method for semantic segmentation called layer-wise training and evaluate it on a light efficient structure called an efficient neural network (ENet). The results of the proposed learning method are compared with the classic learning approaches, including mIoU performance, network robustness to noise, and the possibility of reducing the size of the structure on two RGB image datasets on the road (CamVid) and off-road (Freiburg Forest) paths. Using this method partially eliminates the need for Transfer Learning. It also improves network performance when input is noisy.

Zitieren

Using layer-wise training for Road Semantic Segmentation in Autonomous Cars. / Shashaani, Shahrzad; Teshnehlab, Mohammad; Khodadadian, Amirreza et al.
in: IEEE ACCESS, Jahrgang 11, 10.03.2023, S. 46320 - 46329.

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Shashaani, S, Teshnehlab, M, Khodadadian, A, Parvizi, M, Wick, T & Noii, N 2023, 'Using layer-wise training for Road Semantic Segmentation in Autonomous Cars', IEEE ACCESS, Jg. 11, S. 46320 - 46329. https://doi.org/10.1109/ACCESS.2023.3255988
Shashaani, S., Teshnehlab, M., Khodadadian, A., Parvizi, M., Wick, T., & Noii, N. (2023). Using layer-wise training for Road Semantic Segmentation in Autonomous Cars. IEEE ACCESS, 11, 46320 - 46329. https://doi.org/10.1109/ACCESS.2023.3255988
Shashaani S, Teshnehlab M, Khodadadian A, Parvizi M, Wick T, Noii N. Using layer-wise training for Road Semantic Segmentation in Autonomous Cars. IEEE ACCESS. 2023 Mär 10;11:46320 - 46329. doi: 10.1109/ACCESS.2023.3255988
Shashaani, Shahrzad ; Teshnehlab, Mohammad ; Khodadadian, Amirreza et al. / Using layer-wise training for Road Semantic Segmentation in Autonomous Cars. in: IEEE ACCESS. 2023 ; Jahrgang 11. S. 46320 - 46329.
Download
@article{ffc7c85a44e14e9d90b86d13ee618ff7,
title = "Using layer-wise training for Road Semantic Segmentation in Autonomous Cars",
abstract = "A recently developed application of computer vision is pathfinding in self-driving cars. Semantic scene understanding and semantic segmentation, as subfields of computer vision, are widely used in autonomous driving. Semantic segmentation for pathfinding uses deep learning methods and various large sample datasets to train a proper model. Due to the importance of this task, accurate and robust models should be trained to perform properly in different lighting and weather conditions and in the presence of noisy input data. In this paper, we propose a novel learning method for semantic segmentation called layer-wise training and evaluate it on a light efficient structure called an efficient neural network (ENet). The results of the proposed learning method are compared with the classic learning approaches, including mIoU performance, network robustness to noise, and the possibility of reducing the size of the structure on two RGB image datasets on the road (CamVid) and off-road (Freiburg Forest) paths. Using this method partially eliminates the need for Transfer Learning. It also improves network performance when input is noisy.",
keywords = "Autonomous cars, Computer Vision, Convolution Neural Networks, Layer-wise trains, Semantic Segmentation, layer-wise trains, computer vision, convolution neural networks, semantic segmentation",
author = "Shahrzad Shashaani and Mohammad Teshnehlab and Amirreza Khodadadian and Maryam Parvizi and Thomas Wick and Nima Noii",
note = "Funding Information: The work of Maryam Parvizi was supported by the Alexander von Humboldt Foundation Project.",
year = "2023",
month = mar,
day = "10",
doi = "10.1109/ACCESS.2023.3255988",
language = "English",
volume = "11",
pages = "46320 -- 46329",
journal = "IEEE ACCESS",
issn = "2169-3536",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

Download

TY - JOUR

T1 - Using layer-wise training for Road Semantic Segmentation in Autonomous Cars

AU - Shashaani, Shahrzad

AU - Teshnehlab, Mohammad

AU - Khodadadian, Amirreza

AU - Parvizi, Maryam

AU - Wick, Thomas

AU - Noii, Nima

N1 - Funding Information: The work of Maryam Parvizi was supported by the Alexander von Humboldt Foundation Project.

PY - 2023/3/10

Y1 - 2023/3/10

N2 - A recently developed application of computer vision is pathfinding in self-driving cars. Semantic scene understanding and semantic segmentation, as subfields of computer vision, are widely used in autonomous driving. Semantic segmentation for pathfinding uses deep learning methods and various large sample datasets to train a proper model. Due to the importance of this task, accurate and robust models should be trained to perform properly in different lighting and weather conditions and in the presence of noisy input data. In this paper, we propose a novel learning method for semantic segmentation called layer-wise training and evaluate it on a light efficient structure called an efficient neural network (ENet). The results of the proposed learning method are compared with the classic learning approaches, including mIoU performance, network robustness to noise, and the possibility of reducing the size of the structure on two RGB image datasets on the road (CamVid) and off-road (Freiburg Forest) paths. Using this method partially eliminates the need for Transfer Learning. It also improves network performance when input is noisy.

AB - A recently developed application of computer vision is pathfinding in self-driving cars. Semantic scene understanding and semantic segmentation, as subfields of computer vision, are widely used in autonomous driving. Semantic segmentation for pathfinding uses deep learning methods and various large sample datasets to train a proper model. Due to the importance of this task, accurate and robust models should be trained to perform properly in different lighting and weather conditions and in the presence of noisy input data. In this paper, we propose a novel learning method for semantic segmentation called layer-wise training and evaluate it on a light efficient structure called an efficient neural network (ENet). The results of the proposed learning method are compared with the classic learning approaches, including mIoU performance, network robustness to noise, and the possibility of reducing the size of the structure on two RGB image datasets on the road (CamVid) and off-road (Freiburg Forest) paths. Using this method partially eliminates the need for Transfer Learning. It also improves network performance when input is noisy.

KW - Autonomous cars

KW - Computer Vision

KW - Convolution Neural Networks

KW - Layer-wise trains

KW - Semantic Segmentation

KW - layer-wise trains

KW - computer vision

KW - convolution neural networks

KW - semantic segmentation

UR - http://www.scopus.com/inward/record.url?scp=85149825727&partnerID=8YFLogxK

U2 - 10.1109/ACCESS.2023.3255988

DO - 10.1109/ACCESS.2023.3255988

M3 - Article

AN - SCOPUS:85149825727

VL - 11

SP - 46320

EP - 46329

JO - IEEE ACCESS

JF - IEEE ACCESS

SN - 2169-3536

ER -

Von denselben Autoren