EAR-Net: Efficient Atrous Residual Network for Semantic Segmentation of Street Scenes Based on Deep Learning

被引:8
|
作者
Shin, Seokyong [1 ]
Lee, Sanghun [2 ]
Han, Hyunho [3 ]
机构
[1] Kwangwoon Univ, Dept Plasma Bio Display, 20 Kwangwoon Ro, Seoul 01897, South Korea
[2] Kwangwoon Univ, Ingenium Coll Liberal Arts, 20 Kwangwoon Ro, Seoul 01897, South Korea
[3] Univ Ulsan, Coll Gen Educ, 93 Daehak Ro, Ulsan 44610, South Korea
来源
APPLIED SCIENCES-BASEL | 2021年 / 11卷 / 19期
关键词
atrous spatial pyramid pooling; deep learning; encoder-decoder; residual learning; semantic segmentation; IMAGE;
D O I
10.3390/app11199119
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
Segmentation of street scenes is a key technology in the field of autonomous vehicles. However, conventional segmentation methods achieve low accuracy because of the complexity of street landscapes. Therefore, we propose an efficient atrous residual network (EAR-Net) to improve accuracy while maintaining computation costs. First, we performed feature extraction and restoration, utilizing depthwise separable convolution (DSConv) and interpolation. Compared with conventional methods, DSConv and interpolation significantly reduce computation costs while minimizing performance degradation. Second, we utilized residual learning and atrous spatial pyramid pooling (ASPP) to achieve high accuracy. Residual learning increases the ability to extract context information by preventing the problem of feature and gradient losses. In addition, ASPP extracts additional context information while maintaining the resolution of the feature map. Finally, to alleviate the class imbalance between the image background and objects and to improve learning efficiency, we utilized focal loss. We evaluated EAR-Net on the Cityscapes dataset, which is commonly used for street scene segmentation studies. Experimental results showed that the EAR-Net had better segmentation results and similar computation costs as the conventional methods. We also conducted an ablation study to analyze the contributions of the ASPP and DSConv in the EAR-Net.
引用
收藏
页数:14
相关论文
共 50 条
  • [1] EAR-NET: Error Attention Refining Network For Retinal Vessel Segmentation
    Wang, Jun
    Zhao, Yang
    Qian, Linglong
    Yu, Xiaohan
    Gao, Yongsheng
    2021 INTERNATIONAL CONFERENCE ON DIGITAL IMAGE COMPUTING: TECHNIQUES AND APPLICATIONS (DICTA 2021), 2021, : 161 - 167
  • [2] Complementary Convolution Residual Networks for Semantic Segmentation in Street Scenes with Deep Gaussian CRF
    Li, Yongbo
    Ma, Yuanyuan
    Cai, Wendi
    Xie, Zhongzhao
    Zhao, Tao
    JOURNAL OF ADVANCED COMPUTATIONAL INTELLIGENCE AND INTELLIGENT INFORMATICS, 2021, 25 (01) : 3 - 12
  • [3] A CNN Architecture for Efficient Semantic Segmentation of Street Scenes
    Mazzini, Davide
    Buzzelli, Marco
    Pau, Danilo Pietro
    Schettini, Raimondo
    2018 IEEE 8TH INTERNATIONAL CONFERENCE ON CONSUMER ELECTRONICS - BERLIN (ICCE-BERLIN), 2018,
  • [4] Learning Dilation Factors for Semantic Segmentation of Street Scenes
    He, Yang
    Keuper, Margret
    Schiele, Bernt
    Fritz, Mario
    PATTERN RECOGNITION (GCPR 2017), 2017, 10496 : 41 - 51
  • [5] ACDSSNet: Atrous Convolution-based Deep Semantic Segmentation Network for Efficient Detection of Sickle Cell Anemia
    Das P.K.
    Dash A.
    Meher S.
    IEEE Journal of Biomedical and Health Informatics, 2024, 28 (10) : 1 - 8
  • [6] Full-Resolution Residual Networks for Semantic Segmentation in Street Scenes
    Pohlen, Tobias
    Hermans, Alexander
    Mathias, Markus
    Leibe, Bastian
    30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, : 3309 - 3318
  • [7] Atrous residual convolutional neural network based on U-Net for retinal vessel segmentation
    Wu, Jin
    Liu, Yong
    Zhu, Yuanpei
    Li, Zun
    PLOS ONE, 2022, 17 (08):
  • [8] Deep Residual Coalesced Convolutional Network for Efficient Semantic Road Segmentation
    Ardiyanto, Igi
    Adji, Teguh Bharata
    PROCEEDINGS OF THE FIFTEENTH IAPR INTERNATIONAL CONFERENCE ON MACHINE VISION APPLICATIONS - MVA2017, 2017, : 378 - 381
  • [9] Deep Multi-Resolution Network for Real-Time Semantic Segmentation in Street Scenes
    Wang, Yalun
    Chen, Shidong
    Bian, Huicong
    Li, Weixiao
    Lu, Qin
    2023 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, IJCNN, 2023,
  • [10] Deep Learning-Based Frameworks for Semantic Segmentation of Road Scenes
    Alokasi, Haneen
    Ahmad, Muhammad Bilal
    ELECTRONICS, 2022, 11 (12)