基于bp神经网络的车牌识别技术.pdf-资料库

qq_37795576-11065122-4744300845394926881.pdf-第1页.png

第1页 / 共64页

qq_37795576-11065122-4744300845394926881.pdf-第2页.png

第2页 / 共64页

qq_37795576-11065122-4744300845394926881.pdf-第3页.png

第3页 / 共64页

qq_37795576-11065122-4744300845394926881.pdf-第4页.png

第4页 / 共64页

qq_37795576-11065122-4744300845394926881.pdf-第5页.png

第5页 / 共64页

qq_37795576-11065122-4744300845394926881.pdf-第6页.png

第6页 / 共64页

qq_37795576-11065122-4744300845394926881.pdf-第7页.png

第7页 / 共64页

qq_37795576-11065122-4744300845394926881.pdf-第8页.png

第8页 / 共64页

分类号：密级： U D C：编号：河北工业大学硕士学位论文基于 BP 神经网络的车牌识别技术的研究论文作者：申瑾学生类别：全日制学科门类：工学学科专业：通信与信息系统指导教师：唐红梅职称：副教授

Dissertation Submitted to Hebei University of Technology for The Master Degree of Communition and Information Systems THE RESEARCH ON TECHNIQUES OF CARLICENSE PLATE RECOGNITION BASED ON BP NEURAL NETWORK by Shen Jin Supervisor: Prof. Tang Hongmei December 2013

摘要随着现在经济的快速发展，智能交通系统（ITS）中的车牌识别技术（LPR）也随之进入了更高层次的发展阶段，这项技术被广泛地应用在对车辆管理的多种场合，并在现代交通控制系统领域中占有一席之位。在本课题的研究中主要运用了数字图像处理以及 BP 神经网络等相关计算机处理技术，对车牌识别系统中的预处理模块、车牌定位模块以及车牌字符分割和识别模块的一些相关内容进行了研究分析，并运用 MATLAB 仿真软件进行了模拟仿真。首先，针对现在城市空气污染严重，雾霾天气频发的情况，在系统的预处理阶段对采集到的含雾图像进行了图像去雾、灰度变换及图像滤波等处理。本课题对图像去雾进行了重点论述分析，以暗原色先验为基础，对基于暗原色先验去雾算法进行了系统的研究，并提出了基于暗原色的改进的去雾算法，其核心思想是对其透射率估计的改进，通过实验表明改进后的算法在解决去雾方面的问题上要优于原始算法，尤其在白色及明亮区域能够真实的还原图像的原本颜色特征。其次，在系统的车牌定位和字符分割模块中，分析了常用方法对对车牌区域进行定位和分割存在的缺点及不足，通过研究采用了基于水平扫描、垂直投影的方法对图像中车牌的区域进行了准确定位，并利用车牌区域的先验信息，选取了局部垂直投影法，对车牌的字符串进行了字符分割处理，并将处理后的字符送入车牌识别模块进行字符识别。最后，论述了 BP 神经网络的相关理论及其算法流程，并针对我国车牌的特殊性设计了相应的汉字网络、字母网络和数字字母网络，同时对各个网络中参数的选取进行了分析研究，对设计后的网络进行了训练，实验表明所设计的识别网络不仅保持了原始 BP 神经网络的非线性映射及自学习、自适应能力，而且增强了网络的泛化能力和容错能力，提高了网络的识别正确率。关键词：智能交通系统暗原色车牌识别 BP 神经网络 I

ABSTRACT With the rapid development of economy, the License Plate Recognition(LPR) of the Intelligent Transport System(ITS) has entered higher development stage. It has been widely applied in various occasions related to vehicle management, and occupies a place in the field of traffic control system. The technology of digital image processing, BP neural network has been used in our research work. Same modules related to image preprocessing, license locating, character segmentation and character recognition in license plate recognition system have been studied. The experiments have been done by MATLAB simulation software. Firstly, considering the serious air pollution and haze weather in the city, some preprocessing operation have been done to the collected foggy images including the image defogging, grey level transformation and image filtering. Emphasis is placed on the image defogging in this part. The algorithm of image defogging based on the prior dark channel has been studied systematically. A modified algorithm has been proposed. The core idea of the modified algorithm is the improvement of the transmission rate. The experiment results indicate that the improved algorithm is better than the original on the image defogging, especially in the regions of the white and bright. The original characteristics of the colors in these regions can be restored. Secondly, the shortcoming and deficiencies of the common method in the phase of license plate locating and segmentation have been explained in the corresponding modules. The horizon scans and vertical projection are combined to locate the license plate. The local vertical projection method has been applied to segment the characters by the aid of prior information of the license plate area. And the individual characters have been sent into the recognition module for the character recognition. Finally, the related theories and algorithm process of the BP neural network have been introduced. In consideration of the particularity of the license plate in our country, the corresponding neural networks letter network and letter-number network have been designed. The selections of the parameters in each network have been studied. And the three networks also have been trained. The experiment insults show that the designed networks not only maintain the nonlinear mapping, including Chinese network, III

self-learning and adaptive ability of normal BP neural network, and also enhance the generalization and fault tolerance ability. The recognition accuracy of the designed network has been improved. Keyword：intelligent transport system dark channel license plate recognition BP neural network IV

目录第一章绪论 ................................................................................................................................... - 1 - 1.1 课题背景 .................................................................................................................................... - 1 - 1.2 国内外车牌识别的研究现状 ................................................................................................ - 1 - 1.3 我国汽车牌照的特殊性 ......................................................................................................... - 2 - 1.4 车牌识别系统的广泛应用 .................................................................................................... - 3 - 1.5 本文的主要工作 ....................................................................................................................... - 3 - 第二章图像的预处理 ................................................................................................................. - 7 - 2.1 图像去雾及常用算法 ............................................................................................................. - 7 - 2.2 基于暗原色的图像去雾 ......................................................................................................... - 9 - 2.3 基于暗原色的图像去雾算法改进 .................................................................................... - 11 - 2.3.1 改进算法的思想 ........................................................................................................ - 11 - 2.3.2 改进算法的实现过程 ............................................................................................... - 12 - 2.3.3 改进算法的去雾结果及分析 ................................................................................. - 12 - 2.4 预处理其他操作 .................................................................................................................... - 14 - 2.4.1 图像灰度化 ................................................................................................................. - 14 - 2.4.2 图像拉伸 ..................................................................................................................... - 15 - 2.4.3 图像滤波 ..................................................................................................................... - 15 - 2.5 本章小结 .................................................................................................................................. - 16 - 第三章车牌定位 ........................................................................................................................ - 17 - 3.1 车牌区域特征 ......................................................................................................................... - 17 - 3.2 本文车牌定位算法 ................................................................................................................ - 18 - 3.2.1 基于行扫描的车牌水平定位 ................................................................................. - 18 - 3.2.2 基于投影法的车牌垂直定位 ................................................................................. - 19 - 3.3 本章小结 .................................................................................................................................. - 20 - 第四章车牌字符分割 ............................................................................................................... - 21 - 4.1 字符分割预处理 .................................................................................................................... - 21 - 4.1.1 车牌底色统一 ............................................................................................................ - 21 - 4.1.2 倾斜矫正 ..................................................................................................................... - 22 - 4.1.3 车牌边框的去除 ........................................................................................................ - 23 - 4.2 车牌字符分割 ......................................................................................................................... - 25 - V

资料库

基于bp神经网络的车牌识别技术.pdf

相关推荐

课程资源

热门标签

最新资料