실제도로 상황에서 탐지 정확도 개선을 위해 공간적 집중을 사용하는 차량 번호판 탐지 기법
A Vehicle License Plate Detection Scheme Using Spatial Attentions for Improving Detection Accuracy in Real-Road Situations
0
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
본 논문에서는 실제 도로의 다양한 상황에서도 차량 번호판을 정확하게 탐지하기 위해 차량 번호판의 후보 지역을 공간 집중 영역으로 사용하는 차량 번호판 탐지 모델을 제안하였다. 먼저, 기존의 WPOD-NET이 전처리 과정에서 검출된 차량 영역을 이용하기 때문에 넓은 탐지 후보 영역으로 인해 불필요한 노이즈가 포함되어 탐지 정확도가 낮아짐을 확인하였다. 이를 개선하기 위해차량 번호판의 후보 지역을 공간 집중 영역으로 사용하는 차량 번호판 탐지 모델을 제안하였고, 제안한 방법이 기존 WPOD-NET보다 탐지 정확도를 어느 정도 개선하는지 분석하기 위해 GT 데이터를 기반으로 최적의 공간 집중 영역을 설정한 경우와 함께 탐지 정확도를 비교하였다. 실험에 따르면 제안된 모델이 기존 WPOD-NET에 비해 타이트한 탐지 후보 영역을 갖기 때문에 약20% 더 높은 탐지 정확도를 보임을 확인하였다.
Abstract
In this paper, a vehicle license plate detection scheme is proposed that uses the spatial attention areas to detect accurately the license plates in various real-road situations. First, the previous WPOD-NET was analyzed, and its detection accuracy is evaluated as lower due to the unnecessary noises in the wide detection candidate areas. To resolve this problem, a vehicle license plate detection model is proposed that uses the candidate area of the license plate as a spatial attention areas. And we compared its performance to that of the WPOD-NET, together with the case of using the optimal spatial attention areas using the ground truth data. The experimental results show that the proposed model has about 20% higher detection accuracy than the original WPOD-NET since the proposed scheme uses tight detection candidate areas.
참고문헌 (16)
- K. M. Sajjad, "Automatic License Plate Recognition using Python and OpenCV", Department of Computer Science and Engineering, pp.2010
- Shan Du, "Automatic License Plate Recognition (ALPR): A State-of-the-Art Review", IEEE Transactions on Circuits and Systems for Video Technology, 23(2), pp.311~325 (2013), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- B. Li, "A vehicle license plate recognition system based on analysis of maximally stable extremal regions", Proceedings of 9th IEEE International Conference on Networking, Sensing and Control, pp.399~404 (2012)
- Hendry, "Automatic License Plate Recognition via sliding-window darknet-YOLO deep learning", Image and Vision Computing, 87, pp.47~56 (2019), Elsevier BV[DOI]
- R. Laroca, "A robust real-time automatic license plate recognition based on the Yolo detector", 2018 International Joint Conference on Neural Networks (IJCNN), pp.1~10 (2018)
- "OpenALPR(Automatic License Plate Recognition)"[URL]
- Gabriel Resende Gonçalves, "Benchmark for license plate character segmentation", Journal of Electronic Imaging, 25(5), pp.053034~ (2016), SPIE-Intl Soc Optical Eng[DOI]
- Gee-Sern Hsu, "Application-Oriented License Plate Recognition", IEEE Transactions on Vehicular Technology, 62(2), pp.552~561 (2013), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- S. M. Silva, "License Plate Detection and Recognition in Unconstrained Scenarios", Computer Vision - ECCV 2018, Lecture Notes in Computer Science, 11216 (2018), Springer
- J. Redmon, "You Only Look Once: Unified, Real-Time Object Detection", 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.779~788 (2016)
- S.W. Lee, "A Progressing Direction for Vehicle License Plate Detectors Based on Performance Evaluations Using Emulated Images for Various Real-Road Situations", Video session presented at the 2020 Asia-pacific Society of Convergent Research Interchange 4th Domestic and International Conference (2020)
- G. Bradski, "The OpenCV Library", Dr. Dobb’s Journal of Software Tools (2000)
- J. Redmon, "Yolo9000 : Better, Faster, Stronger", 2017IEEE Conference on Computer Vision and Pattern Recognition(CVPR), pp.6517~6525 (2017)
- Yolov3: An Incremental Improvement, "J. Redmon", ~ (2018)
- Shaoqing Ren, "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks", IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(6), pp.1137~1149 (2017), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- A. Bochkovskiy, "Yolov4 : Optimal speed and accuracy of object detection", arXiv preprint arXiv:2004.10934, ~ (2020)