DOI QR코드

DOI QR Code

An Improved License Plate Recognition Technique in Outdoor Image

옥외영상의 개선된 차량번호판 인식기술

  • Kim, Byeong-jun (Department of Computer Science and Engineering, Chonbuk National University) ;
  • Kim, Dong-hoon (Department of Computer Science and Engineering, Chonbuk National University) ;
  • Lee, Joonwhoan (Department of Computer Science and Engineering, Chonbuk National University)
  • 김병준 (전북대학교 전자정보공학부(컴퓨터공학)) ;
  • 김동훈 (전북대학교 컴퓨터공학부) ;
  • 이준환 (전북대학교 컴퓨터공학부)
  • Received : 2016.08.11
  • Accepted : 2016.10.21
  • Published : 2016.10.25

Abstract

In general LPR(License Plate Recognition) in outdoor image is not so simple differently from in the image captured from manmade environment, because of geometric shape distortion and large illumination changes. this paper proposes three techniques for LPR in outdoor images captured from CCTV. At first, a serially connected multi-stage Adaboost LP detector is proposed, in which different complementary features are used. In the proposed detector the performance is increased by the Haar-like Adaboost LP detector consecutively connected to the MB-LBP based one in serial manner. In addition the technique is proposed that makes image processing easy by the prior determination of LP type, after correction of geometric distortion of LP image. The technique is more efficient than the processing the whole LP image without knowledge of LP type in that we can take the appropriate color to gray conversion, accurate location for separation of text/numeric character sub-images, and proper parameter selection for image processing. In the proposed technique we use DBN(Deep Belief Network) to achieve a robust character recognition against stroke loss and geometric distortion like slant due to the incomplete image processing.

일반적으로 옥외영상에서의 자동차 번호판 인식은 인위적인 환경에서와는 다르게 기하학적으로 왜곡되어 있을 뿐만 아니라 조명 변화도 크기 때문에 단순환 문제가 아니다. 본 논문에서는 일반 CCTV 카메라로 옥외에서 촬영된 영상에서 자동차 번호판 인식을 위한 개선된 기술들을 제안한다. 먼저 다양한 특징을 상보적으로 사용하는 직렬구조의 다단계 Adaboost 검출기를 제안한다. 제안하는 검출기는 MB-LBP 및 Haar-like 특징을 사용하는 Adaboost 구조를 직렬로 연결하여 번호판 검출의 검출성능을 향상시켰다. 또한 검출된 번호판의 기하학적 왜곡을 보정하고 번호판의 타입을 먼저 결정하여 영상처리를 용이하게 하는 방법을 제안한다. 이런 방법은 그래이 변환, 문자/숫자 분리, 분리된 영상의 영상처리 등에서 사전지식 없이 전체 번호판 영상을 이용하는 경우보다 효율적이다. 본 논문에서 DBN(Deep Belief Network)를 문자/숫자 인식기로 사용하여 영상처리과정에서 기인한 획 손실이나 기울어짐 같은 기하학적인 왜곡에서도 강건한 인식률을 달성하였다.

Keywords

References

  1. X. H. Huaifeng Zhang, Wenjing Jia and Q. Wu. "Learningbased license plate detection using global and local features," 18th International Conference on Pattern Recognition, vol. 02, pp. 1102-1105, 2006.
  2. Moon-Yong Jin, "Real-Time Vehicle License Plate Recognition System Using Adaptive Heuristic Segmentation Algorithm", Chonbuk National University, 2013.
  3. Byung-Gil Han, Jong Taek Lee, Kil-Taek Lim, and Yunsu Chung, "Real-time License Plate Detection in High- Resolution Videos using Fastest Cascade Classifier and Core Patterns," Available http://dx.doi.org/10.4218/etrij.15.2314.0077, [Accessed : July 17, 2016]
  4. Tran Duc Duan, Tran Le Hong Du, Tran Vinh Phuoc, Nguyen Viet Hoang, "Building an Automatic Vehicle License-Plate Recognition System,"Intl. Conf. in Computer Science, vol. 05, pp. 21-24, 2005,
  5. Wenjing Jia, Huaifeng Zhang, Xiang-jian He, "Region-based license plate detection", Journal of Network and Computer Applications, vol.30, pp.1324-1333, 2007. https://doi.org/10.1016/j.jnca.2006.09.010
  6. S.-H. Y. Jun-Wei Hsieh and Y.-S. Chen. "Morphology-based license plate detection from complex scenes," the International Conference on Pattern Recognition, vol. 16, pp. 176-180, 2002.
  7. Jin-Ho Kim, "Vehicle License Plate Recognition for Smart Tolling by Selective Sharpening", The Journal of the Korea Contents Association, vol. 14, no. 12, pp. 1-9, 2014.
  8. Jae-Ho Kim, Dong-Jung Kang, "An Ensemble Classifier Based Method to Select Optimal Image Features for License Plate Recognition", KIEE international transactions on power engineering, vol. 65,no. 1, pp.142-149, 2016.
  9. Seon-Hwan Kim, Sung-Kwon Oh, "RBFNNs-based Recognition System of Vehicle License Plate Using Distortion Correction and Local Binarization", KIEE international transactions on power engineering, vol. 65, no. 9, pp.1531-1540, 2016.
  10. GE Hinton, S Osindero, YW The, "A fast learning algorithm for deep belief network", Neural computation MIT Press, 2006.
  11. JAK Suykens, J Vandewalle, "Least squares support vector machine classifiers", Neural processing letters, 1999.
  12. Hungwen Li, Mark A Lavin, Ronald J Le Master, "Fast Hough transform: A hierarchical approach", Computer Vision, Graphics, and Image Processing, vol.36, pp.139-161, 1986. https://doi.org/10.1016/0734-189X(86)90073-3
  13. M Welling, "Fisher linear discriminant analysis", Department of Computer Science, University of Toronto, 2005.
  14. Shaoqing Ren, Kaiming He, Ross Girshock, Jian Sun, "Faster R-CNN: Toward Real-Time Object Detection with Region Proposal Networks", Advances in Neural Information Processing Systems, vol. 28. 2015.
  15. Dong-Hoon Kim, Byeong-Jun Kim, Joonwhoan-Lee "A Cascaded Detector for Vehicle Number Plate from Outdoor Images", The Korean Institute of Communications and Information Sciences, vol. 11, pp.71-72, 2015.