A deep learning-based semantic segmentation architecture for autonomous driving applications
In recent years, the development of smart transportation has accelerated research on semantic segmentation as it is one of the most important problems in this area. A large receptive field has always been the center of focus when designing convolutional neural networks for semantic segmentation. A m...
| Main Authors: | , , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/161373 |
| _version_ | 1811680950498623488 |
|---|---|
| author | Masood, Sharjeel Ahmed, Fawad Alsuhibany, Suliman A. Ghadi, Yazeed Yasin Siyal, M. Y. Kumar, Harish Khan, Khyber Ahmad, Jawad |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Masood, Sharjeel Ahmed, Fawad Alsuhibany, Suliman A. Ghadi, Yazeed Yasin Siyal, M. Y. Kumar, Harish Khan, Khyber Ahmad, Jawad |
| author_sort | Masood, Sharjeel |
| collection | NTU |
| description | In recent years, the development of smart transportation has accelerated research on semantic segmentation as it is one of the most important problems in this area. A large receptive field has always been the center of focus when designing convolutional neural networks for semantic segmentation. A majority of recent techniques have used maxpooling to increase the receptive field of a network at an expense of decreasing its spatial resolution. Although this idea has shown improved results in object detection applications, however, when it comes to semantic segmentation, a high spatial resolution also needs to be considered. To address this issue, a new deep learning model, the M-Net is proposed in this paper which satisfies both high spatial resolution and a large enough receptive field while keeping the size of the model to a minimum. The proposed network is based on an encoder-decoder architecture. The encoder uses atrous convolution to encode the features at full resolution, and instead of using heavy transposed convolution, the decoder consists of a multipath feature extraction module that can extract multiscale context information from the encoded features. The experimental results reported in the paper demonstrate the viability of the proposed scheme. |
| first_indexed | 2024-10-01T03:33:11Z |
| format | Journal Article |
| id | ntu-10356/161373 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T03:33:11Z |
| publishDate | 2022 |
| record_format | dspace |
| spelling | ntu-10356/1613732022-08-30T03:03:35Z A deep learning-based semantic segmentation architecture for autonomous driving applications Masood, Sharjeel Ahmed, Fawad Alsuhibany, Suliman A. Ghadi, Yazeed Yasin Siyal, M. Y. Kumar, Harish Khan, Khyber Ahmad, Jawad School of Electrical and Electronic Engineering Engineering::Electrical and electronic engineering Autonomous Driving Convolutional Neural Network In recent years, the development of smart transportation has accelerated research on semantic segmentation as it is one of the most important problems in this area. A large receptive field has always been the center of focus when designing convolutional neural networks for semantic segmentation. A majority of recent techniques have used maxpooling to increase the receptive field of a network at an expense of decreasing its spatial resolution. Although this idea has shown improved results in object detection applications, however, when it comes to semantic segmentation, a high spatial resolution also needs to be considered. To address this issue, a new deep learning model, the M-Net is proposed in this paper which satisfies both high spatial resolution and a large enough receptive field while keeping the size of the model to a minimum. The proposed network is based on an encoder-decoder architecture. The encoder uses atrous convolution to encode the features at full resolution, and instead of using heavy transposed convolution, the decoder consists of a multipath feature extraction module that can extract multiscale context information from the encoded features. The experimental results reported in the paper demonstrate the viability of the proposed scheme. Published version One of the authors (Harish Kumar) extends his gratitude to the Deanship of Scientific Research at King Khalid University for funding this work through research groups program under grant number R. G. P. 2/198/43. 2022-08-30T03:03:34Z 2022-08-30T03:03:34Z 2022 Journal Article Masood, S., Ahmed, F., Alsuhibany, S. A., Ghadi, Y. Y., Siyal, M. Y., Kumar, H., Khan, K. & Ahmad, J. (2022). A deep learning-based semantic segmentation architecture for autonomous driving applications. Wireless Communications and Mobile Computing, 2022, 1-12. https://dx.doi.org/10.1155/2022/8684138 1530-8669 https://hdl.handle.net/10356/161373 10.1155/2022/8684138 2-s2.0-85133156016 2022 1 12 en Wireless Communications and Mobile Computing © 2022 Sharjeel Masood et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. application/pdf |
| spellingShingle | Engineering::Electrical and electronic engineering Autonomous Driving Convolutional Neural Network Masood, Sharjeel Ahmed, Fawad Alsuhibany, Suliman A. Ghadi, Yazeed Yasin Siyal, M. Y. Kumar, Harish Khan, Khyber Ahmad, Jawad A deep learning-based semantic segmentation architecture for autonomous driving applications |
| title | A deep learning-based semantic segmentation architecture for autonomous driving applications |
| title_full | A deep learning-based semantic segmentation architecture for autonomous driving applications |
| title_fullStr | A deep learning-based semantic segmentation architecture for autonomous driving applications |
| title_full_unstemmed | A deep learning-based semantic segmentation architecture for autonomous driving applications |
| title_short | A deep learning-based semantic segmentation architecture for autonomous driving applications |
| title_sort | deep learning based semantic segmentation architecture for autonomous driving applications |
| topic | Engineering::Electrical and electronic engineering Autonomous Driving Convolutional Neural Network |
| url | https://hdl.handle.net/10356/161373 |
| work_keys_str_mv | AT masoodsharjeel adeeplearningbasedsemanticsegmentationarchitectureforautonomousdrivingapplications AT ahmedfawad adeeplearningbasedsemanticsegmentationarchitectureforautonomousdrivingapplications AT alsuhibanysulimana adeeplearningbasedsemanticsegmentationarchitectureforautonomousdrivingapplications AT ghadiyazeedyasin adeeplearningbasedsemanticsegmentationarchitectureforautonomousdrivingapplications AT siyalmy adeeplearningbasedsemanticsegmentationarchitectureforautonomousdrivingapplications AT kumarharish adeeplearningbasedsemanticsegmentationarchitectureforautonomousdrivingapplications AT khankhyber adeeplearningbasedsemanticsegmentationarchitectureforautonomousdrivingapplications AT ahmadjawad adeeplearningbasedsemanticsegmentationarchitectureforautonomousdrivingapplications AT masoodsharjeel deeplearningbasedsemanticsegmentationarchitectureforautonomousdrivingapplications AT ahmedfawad deeplearningbasedsemanticsegmentationarchitectureforautonomousdrivingapplications AT alsuhibanysulimana deeplearningbasedsemanticsegmentationarchitectureforautonomousdrivingapplications AT ghadiyazeedyasin deeplearningbasedsemanticsegmentationarchitectureforautonomousdrivingapplications AT siyalmy deeplearningbasedsemanticsegmentationarchitectureforautonomousdrivingapplications AT kumarharish deeplearningbasedsemanticsegmentationarchitectureforautonomousdrivingapplications AT khankhyber deeplearningbasedsemanticsegmentationarchitectureforautonomousdrivingapplications AT ahmadjawad deeplearningbasedsemanticsegmentationarchitectureforautonomousdrivingapplications |