Transfer Learning for Classification of Parking Spots using Residual Networks

被引：6
作者：
Gregor, Michal [1 ]
Pirnik, Rastislav [1 ]
Nemec, Dusan [1 ]
机构：
[1] Univ Zilina, Dept Control & Informat Syst, Univ 8215-1, Zilina 01026, Slovakia
来源：
13TH INTERNATIONAL SCIENTIFIC CONFERENCE ON SUSTAINABLE, MODERN AND SAFE TRANSPORT (TRANSCOM 2019) | 2019年 / 40卷
关键词：
deep learning; convolutional networks; parking lots; occupancy detection; residual architecture;
D O I：
10.1016/j.trpro.2019.07.184
中图分类号：
T [工业技术];
学科分类号：
08 ;
摘要：
The paper proposes a classifier with a residual convolutional architecture for visual parking spot classification into classes "empty" and "occupied". The classifier is trained on the well-known PKLot dataset. Transfer of the resulting model to data with new challenging modalities (such as snow, partially obscured vision, reflections, mist, ...) is tested - to this end a new dataset has been collected by the authors. It is shown that the original classifier fails in some of these unfamiliar settings, but that the failure modes can successfully be corrected using transfer learning. (C) 2019 The Authors. Published by Elsevier B.V.
引用
收藏
页码：1327 / 1334
页数：8
相关论文
共 50 条
  • [11] A Transfer Learning Approach for Diabetic Retinopathy Classification Using Deep Convolutional Neural Networks
    Krishnan, Arvind Sai
    Clive, Derik R.
    Bhat, Vilas
    Ramteke, Pravin Bhaskar
    Koolagudi, Shashidhar G.
    IEEE INDICON: 15TH IEEE INDIA COUNCIL INTERNATIONAL CONFERENCE, 2018,
  • [12] Breast cancer masses classification using deep convolutional neural networks and transfer learning
    Shayma’a A. Hassan
    Mohammed S. Sayed
    Mahmoud I Abdalla
    Mohsen A. Rashwan
    Multimedia Tools and Applications, 2020, 79 : 30735 - 30768
  • [13] Breast cancer masses classification using deep convolutional neural networks and transfer learning
    Hassan, Shayma'a A.
    Sayed, Mohammed S.
    Abdalla, Mahmoud, I
    Rashwan, Mohsen A.
    MULTIMEDIA TOOLS AND APPLICATIONS, 2020, 79 (41-42) : 30735 - 30768
  • [14] Leakage Identification of Underground Structures Using Classification Deep Neural Networks and Transfer Learning
    Wang, Wenyang
    Chen, Qingwei
    Shen, Yongjiang
    Xiang, Zhengliang
    SENSORS, 2024, 24 (17)
  • [15] Automatic Modulation Classification Using Involution Enabled Residual Networks
    Zhang, Hao
    Yuan, Lu
    Wu, Guangyu
    Zhou, Fuhui
    Wu, Qihui
    IEEE WIRELESS COMMUNICATIONS LETTERS, 2021, 10 (11) : 2417 - 2420
  • [16] Adventitious Respiratory Classification using Attentive Residual Neural Networks
    Yang, Zijiang
    Liu, Shuo
    Song, Meishu
    Parada-Cabaleiro, Emilia
    Schuller, Bjoern W.
    INTERSPEECH 2020, 2020, : 2912 - 2916
  • [17] Radio Modulation Classification Using Deep Residual Neural Networks
    Abbas, Adeeb
    Pano, Vasil
    Mainland, Geoffrey
    Dandekar, Kapil
    2022 IEEE MILITARY COMMUNICATIONS CONFERENCE (MILCOM), 2022,
  • [18] ECG Classification using Deep Transfer Learning
    Gajendran, Mohan Kumar
    Khan, Muhammad Zubair
    Khattak, Muazzam A. Khan
    2021 4TH INTERNATIONAL CONFERENCE ON INFORMATION AND COMPUTER TECHNOLOGIES (ICICT 2021), 2021, : 1 - 5
  • [19] ISAR Ship Classification Using Transfer Learning
    Zhao, Weiliang
    Heng, Anthony
    Rosenberg, Luke
    Nguyen, Si Tran
    Hamey, Len
    Orgun, Mehmet
    2022 IEEE RADAR CONFERENCE (RADARCONF'22), 2022,
  • [20] Automatic crack classification and segmentation on masonry surfaces using convolutional neural networks and transfer learning
    Dais, Dimitris
    Bal, Ihsan Engin
    Smyrou, Eleni
    Sarhosis, Vasilis
    AUTOMATION IN CONSTRUCTION, 2021, 125
12345