基于Gabor特征和协同神经网络的车牌识别方法

doi:10.3969/j.issn.1000-1565.2016.02.017

摘要/Abstract

摘要： 研究了车牌字符识别问题,针对车牌识别系统易受天气及光照变化影响的实际应用,将Gabor特征和协同神经网络应用在车牌字符识别中,提高了识别率.首先对车牌字符进行二值化和切分,然后利用Gabor滤波器提取车牌字符的特征参数;再利用协同模式训练特征参数,进而得出训练样本;最后根据协同神经网络进一步识别车牌字符.通过大量仿真实验表明,该方法在不同场景、光照条件下,与传统方法相比,识别率有了较大改进,该方法在车牌识别领域有较强的实用性.

关键词: 特征提取, 神经网络, 车牌识别, 字符分割, Gabor变换

Abstract: Vehicle license recognition has been studied under the climate and light conditions.The feature extraction of Gabor and synergetic neural network has been used to enhance the recognition rate.First,the characters of vehicle license go through the binarization and segmenting to extract the characteristic parameters of the license characters by Gabor filter.Then,the synergetic mode is used to train the characteristic parameters and work out the training sample.Finally,the synergetic neural network is utilized to recognize the license character.Under a variety of environments and light conditions, this approach achieves a much higher rate of recognition rate compared with the traditional mode suggesting that this new method is very effective in the field of vehicle license recognition.

Key words: Feature extraction, Neural network, license plate recognition, character segmentation, Gabor filter

中图分类号:

TP391

石贵民,余文森,肖钟捷. 基于Gabor特征和协同神经网络的车牌识别方法[J]. 河北大学学报(自然科学版), 2016, 36(2): 210-217.

SHI Guimin,YU Wensen,XIAO Zhongjie. A method for license plate recognition in vehicle based on Gabor feature and synergetic neural network[J]. Journal of Hebei University (Natural Science Edition), 2016, 36(2): 210-217.

参考文献

[1] 王敏,黄心汉,魏武,等.一种模板匹配和神经网络的车牌字符识别方法[J].华中科技大学学报(自然科学版),2001,29(03):48-50.DOI:10.3321/j.issn:1671-4512.2001.03.017. WANG Min,HUANG Xinhan,WEI Wu,et al.A vehicle license character recognition method based on module matching and neural network [J].Journal of Huazhong University of Science and Technology,2001,29(3):48-50.DOI:10.3321/j.issn:1671-4512.2001.03.017.
[2] 任柯昱,唐丹,尹显东.基于字符结构知识的车牌汉字快速识别技术[J].计算机测量与控制,2005,13(6):592-594.DOI:10.3321/j.issn:1671-4598.2005.06.031. REN Keyu,TANG Dan,YIN Xiandong.Fast recognition method of license plate character based on image structure feature analysis [J].Computer Measurement & Control,2005,13(6):592-594.DOI:10.3321/j.issn:1671-4598.2005.06.031.
[3] 郭荣艳.BP神经网络在车牌字符识别中的应用研究[J].计算机仿真,2010,27(9):299-301,350.DOI:10.3969/j.issn.1006-9348.2010.09.076. GUO Rongyan.The study of the application of neural network in the vehicle license character recognition [J].Computer Simulation,2010,27(9):299-301,350.DOI:10.3969/j.issn.1006-9348.2010.09.076.
[4] ANAGNOSTOPOULOS C,ANAGNOSTOPOULOS I,LOUMOS V,et al.License plate recognition from still images and video sequences:A survey[J].IEEE Transactions on Intelligent Transportation Systems,2008,9(3):377-391.
[5] 权金娟,温显斌,王春东,等.基于两种神经网络的SAR图像多尺度分割与分析[J].光电子·激光,2008,19(4):542-545.DOI:10.3321/j.issn:1005-0086.2008.04.031. QUAN Jinjuan,WEN Xianbin,WANG Chundong.et al.Multiscale segmetation of SAR images via two neural networks[J].Journal of Optoelectronics·Laser,2008,19(4):542-545.DOI:10.3321/j.issn:1005-0086.2008.04.031.
[6] 杨晓敏.基于Gabor变换和支持向量机的车牌字符识别算法[J].四川大学学报(工程科学版),2005,37(5):130-134,138.DOI:10.3969/j.issn.1009-3087.2005.05.026. YANG Xiaomin.The vehicle license character recognition method based on Gabor transformation and support vector machine[J].Journal of Sichuan University(Engineering science version),2005,37(5):130-134,138.DOI:10.3969/j.issn.1009-3087.2005.05.026.
[7] 吕秀丽,谷立冬,祝立群,等.基于数学形态学和垂直投影的车牌定位[J].科学技术与工程,2012,12(10):2465-2467.DOI:10.3969/j.issn.1671-1815.2012.10.051. LÜ Xiuli,GU Lidong,ZHU Liqun,et al.The vehicle license location based on mathematical morphology and vertical projection [J].Scientific technology and engineering,2012,12(10):2465-2467.DOI:10.3969/j.issn.1671-1815.2012.10.051.
[8] 石贵民.基于扫描线的新型车牌定位方法研究[J].科学技术与工程,2014,14(8):218-221.DOI:10.3969/j.issn.1671-1815.2014.08.042. SHI Guimin.Research of a new vehicle license plate location method based on scan line [J].Scientific technology and engineering,2014,14(8):218-221.DOI:10.3969/j.issn.1671-1815.2014.08.042.
[9] 刘松,孙东,卢一项.基于单尺度Retinex雾天降质图像增强算法[J].微电子学与计算机,2013,30(11):43-46,50. LIU Song,SUN Dong,LU Yixiang.The fog-degraded image enhancement algorithm based on a single-dimension Retinex [J].Microelectronics & Compute,2013,30(11):43-46,50.
[10] 万金娥,袁保社,谷朝,等.基于字符归一化双投影互相关性匹配识别算法[J].计算机应用,2013,33(3):645-647,716. WANG Jine,YUAN Baoshe,GU Chao,et al.The double projection correlation match recognition algorithm based on character normalization [J].Computer application,2013,33(3):645-647,716.
[11] 黑光月,王江晴,孙阳光.脱机手写女书字符图像多方向文本行提取[J].计算机应用研究,2013,30(2):627-630.DOI:10.3969/j.issn.1001-3695.2013.02.086. HEI Guangyue,WANG Jiangqing,SUN Yangguang.Multi-oriented text line extraction from offline Nǚshu characters image [J].Computer application,2013,30(2):627-630.DOI:10.3969/j.issn.1001-3695.2013.02.086.
[12] GABOR D.Theory of communication[J].Journal of Institute for Electrical Engineering,1946,93:429-457.
[13] DAUGMAN J G.Two dimensional spectral analysis of cortical receptive field profiles[J].Vision Research,1980,20:847-856.
[14] 谢劲芳.基于Gabor小波变换的人脸表情识别[J].通信技术,2014,47(1):33-36.DOI:10.3969/j.issn.1002-0802.2014.01.007. XIE Jinfang.The facial expression recognition based on Gabor wavelet transform [J].Communications Technology,2014,47(1):33-36.DOI:10.3969/j.issn.1002-0802.2014.01.007.
[15] HAKEN H.Synergetics and introduction[M].Berlin:Spinger,1977.
[16] 刘秉瀚,王伟智,方秀端.协同模式识别方法综述[J].系统工程与电子技术,2003,25(6):758-762.DOI:10.3321/j.issn:1001-506X.2003.06.030. LIU Binghan,WANG Weizhi,FANG Xiuduan.A summary of synergetic pattern recognition methods [J].System engineering and electronic technology,2003,25(6):758-762.DOI:10.3321/j.issn:1001-506X.2003.06.030.
[17] 方秀端.协同模式识别算法的研究与应用[D].福州:福州大学,2001. FANG Xiuduan.The synergetic pattern recognition algorithm and application [D].Fuzhou:Fuzhou University,2001.
[18] 葛强.基于协同学的数字指纹技术[D].武汉:华中科技大学,2007. GE Qiang.Digital fingerprinting based on synergetic[D].Wuhan:Huazhong University of Science and Technology,2007.
[19] 张丽熙.基于纹理特征空间的淋巴组织病理图像协同分类研究[D].福州:福州大学,2006. ZHANG Lixi.The study on lymphoid tissue color pathological images synergetic classification based on the textural feature space[D].Fuzhou:Fuzhou University,2006.
[20] 黄敏霁.离线手写体笔迹鉴别多分类器研究[D].福州:福州大学,2002. HUANG Minji.Offline handwriting identification multi-classifier study [D].Fuzhou:Fuzhou University,2002.