Journal of Broadcast Engineering (방송공학회논문지)

The Korean Institute of Broadcast and Media Engineers (한국방송∙미디어공학회)
권호 표지
DOI QR코드

DOI QR Code

2D Artificial Data Set Construction System for Object Detection and Detection Rate Analysis According to Data Characteristics and Arrangement Structure: Focusing on vehicle License Plate Detection

객체 검출을 위한 2차원 인조데이터 셋 구축 시스템과 데이터 특징 및 배치 구조에 따른 검출률 분석 : 자동차 번호판 검출을 중점으로

  • Kim, Sang Joon (Dept. of Information Technology and Media Engineering, The graduate School of Nano IT Design Fusion, Seoul National University of Science and Technology) ;
  • Choi, Jin Won (Dept. of Mechanical System Design Engineering, Seoul National University of Science and Technology) ;
  • Kim, Do Young (Dept. of Electrical and Information Engineering, Seoul National University of Science and Technology) ;
  • Park, Gooman (Dept. of Electronic IT Media Engineering, Seoul National University of Science and Technology)
  • 김상준 (서울과학기술대학교 정보통신미디어공학전공) ;
  • 최진원 (서울과학기술대학교 기계시스템 디자인공학과) ;
  • 김도영 (서울과학기술대학교 전기정보공학과) ;
  • 박구만 (서울과학기술대학교 전자 IT 미디어공학과)
  • Received : 2022.01.17
  • Accepted : 2022.02.14
  • Published : 2022.03.30
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, deep learning networks with high performance for object recognition are emerging. In the case of object recognition using deep learning, it is important to build a training data set to improve performance. To build a data set, we need to collect and label the images. This process requires a lot of time and manpower. For this reason, open data sets are used. However, there are objects that do not have large open data sets. One of them is data required for license plate detection and recognition. Therefore, in this paper, we propose an artificial license plate generator system that can create large data sets by minimizing images. In addition, the detection rate according to the artificial license plate arrangement structure was analyzed. As a result of the analysis, the best layout structure was FVC_III and B, and the most suitable network was D2Det. Although the artificial data set performance was 2-3% lower than that of the actual data set, the time to build the artificial data was about 11 times faster than the time to build the actual data set, proving that it is a time-efficient data set building system.

최근 객체 인식에 높은 성능을 가진 딥러닝 네트워크가 나오고 있다. 딥러닝을 이용한 객체 인식의 경우 성능 향상을 위해 학습 데이터 셋 구축이 중요하다. 데이터 셋을 구축하기 위해서는 이미지를 수집하고 라벨링 해야 한다. 이 과정은 많은 시간과 인력이 필요하다. 때문에 오픈 데이터 셋을 사용한다. 그러나 방대한 오픈 데이터 셋을 가지고 있지 않는 객체도 존재한다. 그 중 하나가 번호판 검출과 인식에 필요한 데이터이다. 이에 본 논문에서는 이미지를 최소화 하여 대용량 데이터 셋을 만들 수 있는 인조 번호판 생성기 시스템을 제안한다. 또한 인조 번호판 배치구조에 따른 검출률을 분석했다. 분석결과 가장 좋은 배치구조는 FVC_III, B이며 가장 적합한 네트워크는 D2Det이었다. 인조 데이터셋 성능은 실제 데이터셋의 성능보다 2~3%가 낮았지만, 인조 데이터를 구축하는 시간이 실제 데이터셋을 구축하는 시간보다 약 11배 빨라 시간적으로 효율적인 데이터 셋 구축 시스템임을 증명하였다.

Keywords

Acknowledgement

This work was supported by Institute for Information & Communications Technology Planning & Evaluation(IITP) grant funded by the Korea government(MSIT) in 2022(No. 2017-0-00217, Development of Immersive Signage Based on Variable Transparency and Multiple Layers).

References

  1. OpenImage [Internet] https://storage.googleapis.com/openimages/web/ factsfigures.html
  2. MS CoCO [Internet] https://cocodataset.org/#home
  3. CIFAT-10 [Internet] http://www.cs.toronto.edu/~kriz/cifar.html
  4. Jin-Ho Kim, and Duck-soo Noh "Vehicle License Plate Recognition System By Edge-based Segment Image Generation", JOURNAL OF THE KOREA CONTENTS ASSOCIATION 12(3), 2012.3, 9-16 doi: https://doi.org/10.5392/JKCA.2012.12.03.009
  5. Hyoung-chul oh and Jong-Ho choi "A Recognition Algorithm of Car License Plate using Edge Projection and Directivity Vector", The Journal of Korean Institute of Information Technology 7(1), 2009.2, 83-92. https://www.dbpia.co.kr/journal/articleDetail?nodeId=NODE01149744
  6. Seung-ju Lee and Goo-man Park, "Proposal for License Plate Recognition Using Synthetic Data and Vehicle Type Recognition System", JOURNAL OF BROADCAST ENGINEERING 25(5), 2020.9, 776-788. doi: https://doi.org/10.5909/JBE.2020.25.5.776
  7. Jung-Hwan Kim, Joon-Hong Lim, "License Plate Detection and Recognition Algorithm using Deep Learning", journal of IKEEE 23(2), 2019.6, 642-651. doi: http://dx.doi.org/10.7471/ikeee.2019.23.2.642
  8. T. Bjorklund, A. Fiandrotti, M. Annarumma, G. Francini and E. Magli, "Automatic License Plate Recognition with Convolutional Neural Networks Trained on Synthetic Data," International Workshop on Multimedia Signal Processing (MMSP), pp. 1-6, 2017. doi: http://dx.doi.org/10.1109/MMSP.2017.8122260
  9. J. Ha "Detection of Korea License Plate by Mask R-CNN using Composite Image" Journal of Institute of Control, Robotics and Systems 26(9), 2020.9, 778-783. doi: http://dx.doi.org/10.5302/J.ICROS.2020.20.0070
  10. S. Kim, Y. Lee and G. Park, "Artificial license plate generation system for vehicle license plate detection" Korean society internet Information, VOL 21 NO.01 PP.33~34. 2020. https://www.eiric.or.kr/literature/ser_view.php? SnxGubun=IN KO&mode=total&searchCate=literature&gu=INME020E0&cmd=qryview&SnxIndxNum=232898&rownum=&totalCnt=2&rownum=2&q1_t=7J247KGwIOuyiO2YuO2MkA==&listUrl=L3NlYXJjaC9yZXN1bHQucGhwP1NueEd1YnVuPUlOS08mbW9kZT10b3RhbCZzZWFyY2hDYXRlPWxpdGVyYXR1cmUmcTE9JUMwJUNFJUMxJUI2KyVCOSVGOCVDOCVBMyVDNiVDNyZ4PTAmeT0w&q1=%C0%CE%C1%B6+%B9%F8%C8%A3%C6%C7&kci=
  11. Girshick, R., Donahue, J., Darrell, T., Malik, J.: Rich feature hierarchies for accurate object detection and semantic segmentation. arXiv:1311.2524 (2014). doi: https://doi.org/10.48550/arXiv.1311.2524
  12. J. Redmon, S. Divvala, R. Girshick, and A. Farhadi. You only look once: Unified, real-time object detection. arXiv preprint arXiv:1506.02640, 2015. doi: https://doi.org/10.48550/arXiv.1506.02640
  13. Joseph Redmon and Ali Farhadi. YOLOv3: An incremental improvement. arXiv preprint arXiv:1804.02767, 2018. doi: https://doi.org/10.48550/arXiv.1804.02767
  14. W. Liu, D. Anguelov, D. Erhan, C. Szegedy, and S. Reed. Ssd: Single shot multibox detector. arXiv preprint arXiv:1512.02325, 2015. doi: https://doi.org/10.1007/978-3-319-46448-0_2
  15. Qijie Zhao, Tao Sheng, Yongtao Wang, Zhi Tang, Ying Chen, Ling Cai, and Haibin Ling. M2det: A single-shot object detector based on multi-level feature pyramid network. In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), volume 33, pages 9259-9266, 2019. doi: https://doi.org/10.1609/aaai.v33i01.33019259
  16. Zhang, S., Wen, L., Bian, X., Lei, Z., Li, S.Z.: Single-shot refinement neural network for object detection. arXiv preprint arXiv:1711.06897 doi: https://doi.org/10.48550/arXiv.1711.06897
  17. J. Cao, H. Cholakkal, R. M. Anwer, F. S. Khan, Y. Pang and L. Shao, "D2Det: Towards High Quality Object Detection and Instance Segmentation," 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 11482-11491, 2020. doi: https://doi.org/10.1109/CVPR42600.2020.01150
  18. How to Create a synthetic Dataset for computer vision [Internet] https://blog.roboflow.com/how-to-create-a-synthetic-dataset-forcomputer-vision/
  19. S.R Richter, V. Vineet, S. Roth, and V. Koltun "Playing for Data: Ground Truth from Computer Games." European Conference on Computer Vision -ECCV 2016 pp 102-118, 2016. doi: https://doi.org/10.48550/arXiv.1608.02192
  20. CamVid Database [Internet] http://mi.eng.cam.ac.uk/research/projects/VideoRec/
  21. SYNTHIA dataset [Internet] http://synthia-dataset.net/
  22. M. Muller, V. Casser, N. Smith, D.L Michels and B. Ghanem "Teaching UAVs to Race: End-to-End Regression of Agile Controls in Simulation" European Conference on Computer Vision-ECCV 2018 pp 11-29. doi: 2018. https://doi.org/10.48550/arXiv.1708.05884
  23. NAVER LABS KITTI dataset [Internet] https://europe.naverlabs.com/Research/Computer-Vision/Proxy-Virtual-Worlds/
  24. J. Tremblay, et. al. "Training Deep Networks with Synthetic Data: Bridging the Reality Gap by Domain Randomization". Conference on Computer Vision and Pattern Recognition Workshops(CVPRW) 2018. doi: https://doi.org/10.48550/arXiv.1804.06516
  25. Training Deep Networks with Synthetic Data Bridging the Reality Gap by Domain Randomization Review [Internet] https://hoya012.github.io/blog/Tutorials-of-Object-Detection-Using-Deep-Learning-performance-one/