首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于多层前馈神经网络的车牌提取方法
引用本文:刘军杰,张南.基于多层前馈神经网络的车牌提取方法[J].广东交通职业技术学院学报,2007,6(1):27-29,32.
作者姓名:刘军杰  张南
作者单位:1. 广东交通职业技术学院,广东广州,510650;华南师范大学,广东广州,510631
2. 中船重工集团第719研究所,湖北武汉,430068
摘    要:文中提出一种采用三层前馈神经网络提取车牌的方法。该方法是一种基于纹理分析的图像分割算法,适合于彩色及灰度图像。实验表明,该方法能准确地提取图像中的车牌。

关 键 词:车牌提取  多层前馈神经网络  特征提取  图像滤波  阈值化
文章编号:1671-8496-(2007)01-0027-03
修稿时间:2006-05-09

Multilayer Feedforward Neural Network Based Approach to Vehicle License Plate Extraction
LIU Jun-jie,ZHANG Nan.Multilayer Feedforward Neural Network Based Approach to Vehicle License Plate Extraction[J].Journal of Guangdong Communication Polytechnic,2007,6(1):27-29,32.
Authors:LIU Jun-jie  ZHANG Nan
Institution:1 .Guangdong Communications Polytechnic, Guangzhou 510650, China; 2.South China Normal University, Guangzhou 510631, China;3.The 719 Research Institute of China Shipbuilding Industry, Corporation,Wuhan 430068
Abstract:In this paper,the authors present a multilayer feedforward neural network(MFNN) based approach to vehicle license plate extraction.This approach adapts to both color and gray image for the reason that it is a kind of texture-based image segmentation method.Experimental results show that vehicle license plate can be extracted from image correctly using the proposed algorithm.
Keywords:extraction of vehicle license plate  multilayer feedforward neural network  feature extraction  image filtering  thresholding
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号