Car parking detection in images by using a semi-supervised modified YOLOv5 model

Authors

  • Zhang Shuai Luoyang Scorpio Information Technology Ltd., Luoyang 471000, Henan, China; Belarusian State University, 4 Niezaliezhnasci Avenue, Minsk 220030, Belarus
  • Guangdi Ma EarthView Image Inc., 11 Keyuan Road, Huzhou 313200, China
  • Yang Weichen EarthView Image Inc., 11 Keyuan Road, Huzhou 313200, China
  • Fang Zuo Henan University, 85 Minglun Street, Kaifeng 475004, China
  • Sergey V. Ablameyko Belarusian State University, 4 Niezaliezhnasci Avenue, Minsk 220030, Belarus cEarthView Image Inc., 11 Keyuan Road, Huzhou 313200, China; United Institute of Informatics Problems, National Academy of Sciences of Belarus, 6 Surganava Street, Minsk 220012, Belarus

Keywords:

car parking detection, semi-supervised learning, YOLOv5 neural network

Abstract

The problem of car parking detection in images attracts the attention of many researchers. In this task, it is quite difficult to identify rectangular, continuous parking spaces in all kinds of city images under different weather conditions, combining the low-light environment and the system’s low cost with high detection accuracy. In this paper, we propose a modified version of the YOLOv5 model joined with semi-supervised learning that allows us to detect parking lots in any complex scene, independent of parking space lines and parking environments. Due to the combination of the nature of semi-supervised learning and the high accuracy of supervised learning models, the modified version of YOLOv5 model permits to use very little labeled data and a large amount of unlabeled data. It can significantly reduce training time while maintaining recognition accuracy. Compared with other neural network models, the modified version of YOLOv5 model has the characteristics of fast training speed, persistent operation, small model size, and high model precision and recall values.

Author Biographies

  • Zhang Shuai, Luoyang Scorpio Information Technology Ltd., Luoyang 471000, Henan, China; Belarusian State University, 4 Niezaliezhnasci Avenue, Minsk 220030, Belarus

    research and development manager, Luoyang Scorpio Information Technology Ltd., and postgraduate student at the department of web-technologies and computer modelling, faculty of mechanics and mathematics, Belarusian State University

     

  • Guangdi Ma, EarthView Image Inc., 11 Keyuan Road, Huzhou 313200, China

    chief engineer

     

  • Yang Weichen, EarthView Image Inc., 11 Keyuan Road, Huzhou 313200, China

    general manager

     

  • Fang Zuo, Henan University, 85 Minglun Street, Kaifeng 475004, China

    director of Henan International Joint Laboratory of Theories and Key Technologies on Intelligence Networks

     

  • Sergey V. Ablameyko, Belarusian State University, 4 Niezaliezhnasci Avenue, Minsk 220030, Belarus cEarthView Image Inc., 11 Keyuan Road, Huzhou 313200, China; United Institute of Informatics Problems, National Academy of Sciences of Belarus, 6 Surganava Street, Minsk 220012, Belarus

    doctor of science (engineering), academician of the National Academy of Sciences of Belarus, full professor; professor at the department of web-technologies and computer simulation, faculty of mechanics and mathematics, Belarusian State University, and chief researcher at the department of intelligent information systems, United Institute of Informatics Problems, National Academy of Sciences of Belarus

     

References

  1. Alam M, Moroni D, Pieri G, Tampucci M, Gomes M, Fonseca J, et al. Real-time smart parking systems integration in distributed ITS for smart cities. Journal of Advanced Transportation. 2018;2018:1485652. DOI: 10.1155/2018/1485652.
  2. Faheem F, Mahmud SA, Khan GM, Rahman M, Zafar H. A survey of intelligent car parking system. Journal of Applied Research and Technology. 2013;11(5):714–726. DOI: 10.1016/S1665-6423(13)71580-3.
  3. Yusnita R, Norbaya F, Basharuddin N. Intelligent parking space detection system based on image processing. International Journal of Innovation, Management and Technology. 2012;3(3):232–235. DOI: 10.7763/IJIMT.2012.V3.228.
  4. Huang Ching-Chun, Vu Hoang Tran, Chen Yi-Ren. A multiclass boosting approach for integrating weal classifiers in parking space detection. In: 2015 IEEE International conference on consumer electronics; 2015 June 6–8; Taipei, Taiwan. [S. l.]: IEEE; 2015. p. 314–315. DOI: 10.1109/ICCE-TW.2015.7216918.
  5. Bohush R, Yarashevich P, Ablameyko S, Kalganova T. Extraction of image parking spaces in intelligent video surveillance systems. Machine Graphics and Vision. 2018;27(1–4):47–62. DOI: 10.22630/MGV.2018.27.1.3.
  6. Wang Hai, Yu Yljie, Cai Yingfeng, Chen Xiaobo, Chen Long, Liu Qingchao. A comparative study of state-of-the-art deep learning algorithms for vehicle detection. IEEE Intelligent Transportation Systems Magazine. 2019;11(2):82–95. DOI: 10.1109/MITS.2019. 2903518.
  7. Acharya D, Yan W, Khoshelham K. Real-time image-based parking occupancy detection using deep learning. In: Peters S, Khoshelham K, editors. Research@Locate­2018. Proceedings of the 5th annual conference; 2018 April 9–11; Adelaide, Australia. [S. l.]: [s. n.]; 2018. p. 33–40 (CEUR workshop proceedings; volume 2087).
  8. Nyambal J, Klein R. Automated parking space detection using convolutional neural networks. arXiv:2106.07228v1 [Preprint]. 2021 [cited 2023 January 10]: [6 p.]. Available from: https://arxiv.org/abs/2106.07228v1.
  9. Naufal AA, Fatichah C, Suciati N. Preprocessed mask RCNN for parking space detection in smart parking systems. International Journal of Intelligent Engineering and Systems. 2020;13(6):255–265. DOI: 10.22266/ijies2020.1231.23.
  10. Ahmad J, Lewis Z, Duraisamy P, McDonald T. Parking lot monitoring using MRCNN. In: 10 th International conference on computing, communication and networking technologies (ICCCNT); 2019 July 6–8; Kanpur, India. [S. l.]: IEEE; 2019. p. 10–13. DOI: 10.1109/ICCCNT45670.2019.8944394.
  11. Agrawal T, Urolagin S. Multi-angle parking detection system using mask R-CNN. In: BDET­2020. Proceedings of the 2 nd International conference on big data engineering and technology; 2020 January 3–5; Singapore, China. New York: Association for Computing Machinery; 2020. p. 76–80. DOI: 10.1145/3378904.3378914.
  12. Yucheng Guo, Hongtao Shi. Automatic parking system based on improved neural network algorithm and intelligent image analysis. Computational Intelligence and Neuroscience. 2021;2021:4391864. DOI: 10.1155/2021/4391864.
  13. Miao Y, Liu F, Hou T, Liu L, Liu Y. A nighttime vehicle detection method based on YOLOv3. In: 2020 Chinese automation congress (CAC); 2020 November 6–8; Shanghai, China. [S. l.]: IEEE; 2020. p. 6617–6621. DOI: 10.1109/CAC51589. 2020.9326819.
  14. Wang Hai, Lou Xinyu, Cai Yingfeng, Li Yicheng, Chen Long. Real-time vehicle detection algorithm based on vision and lidar point cloud fusion. Journal of Sensors. 2019;2019:8473980. DOI: 10.1155/2019/8473980.
  15. Carrasco DP, Rashwan HA, García MÁ, Puig D. T-YOLO: tiny vehicle detection based on YOLO and multi-scale convolutional neural networks. IEEE Access. 2023;11:22430–22440. DOI: 10.1109/ACCESS.2021.3137638.
  16. Bura H, Lin N, Kumar N, Malekar S, Nagaraj S, Liu K. An edge based smart parking solution using camera networks and deep learning. 2018 IEEE International conference on cognitive computing (ICCC); 2018 July 2–7; San Francisco, USA. [S. l.]: IEEE; 2018. p. 17–24. DOI: 10.1109/ICCC.2018.00010.
  17. Uzar M, Öztürk Ş, Bayrak OC, Arda T, Öcalan NT. Performance analysis of YOLO versions for automatic vehicle detection from UAV images. Advanced Remote Sensing Journal. 2021;1(1):16–30.
  18. Detailed YOLOv5 model framework [Internet]. Beijing: Lezhi Network Technology Co.; 2021 March 3 [updated 2022 July 12; cited 2022 November 19]. Available from: https://blog.csdn.net/qq_16792139/article/details/114310670.
  19. Zhu X, Goldberg AB. Introduction to semi­supervised learning. Cham: Springer; 2009. XII, 116 p. (Brachman RJ, Dietterich T, editors. Synthesis lectures on artificial intelligence and machine learning; lecture 6). DOI: 10.1007/978-3-031-01548-9.
  20. Rosenberg C, Hebert M, Schneiderman H. Semi-supervised self-training of object detection models. In: Seventh IEEE workshops on applications of computer vision (WACV/MOTION’05); 2005 January 5–7; Breckenridge, USA. Volume 1. [S. l.]: IEEE; 2005. p. 29–36. DOI: 10.1109/ACVMOT.2005.107.
  21. Lee Dong-Hyun. Pseudo-label: the simple and efficient semi-supervised learning method for deep neural networks. In: ICML 2013 workshop: challenges in representation learning (WREPL); 2013 June 16–21; Atlanta, USA. Volume 3. [S. l.]: [s. n.]; 2013. p. 2–4.
  22. Sohn K, Zhang Z, Li C-L, Zhang H, Lee C-Y, Pfister T. A simple semi-supervised learning framework for object detection. arXiv:2005.04757 [Preprint]. 2020 [cited 2023 January 21]: [15 p.]. Available from: https://arxiv.org/abs/2005.04757v2.
  23. Janocha K, Czarnecki WM. On loss functions for deep neural networks in classification. arXiv:1702.05659 [Preprint]. 2017 [cited 2023 January 21]: [10 p.]. Available from: https://arxiv.org/abs/1702.05659v1.
  24. Zhang Haoting, Tian Mei, Shao Gaoping, Cheng Juan, Liu Jingjing. Target detection of forward-looking sonar image based on improved YOLOv5. IEEE Access. 2022;10:18023–18034. DOI: 10.1109/ACCESS.2022.3150339.

Downloads

Additional Files

Published

2023-12-22

Issue

No. 3 (2023): Journal of the Belarusian State University. Mathematics and Informatics

Section

Theoretical Foundations of Computer Science

License

The authors who are published in this journal agree to the following:

  1. The authors retain copyright on the work and provide the journal with the right of first publication of the work on condition of license Creative Commons Attribution-NonCommercial. 4.0 International (CC BY-NC 4.0).
  2. The authors retain the right to enter into certain contractual agreements relating to the non-exclusive distribution of the published version of the work (e.g. post it on the institutional repository, publication in the book), with the reference to its original publication in this journal.
  3. The authors have the right to post their work on the Internet (e.g. on the institutional store or personal website) prior to and during the review process, conducted by the journal, as this may lead to a productive discussion and a large number of references to this work. (See The Effect of Open Access.)

How to Cite

[1]
Shuai, Z. et al. 2023. Car parking detection in images by using a semi-supervised modified YOLOv5 model. Journal of the Belarusian State University. Mathematics and Informatics. 3 (Dec. 2023), 72–81.