首页 | 本学科首页   官方微博 | 高级检索  
     

基于不同隐马尔科夫模型的图像识别方法
引用本文:李楠,姬光荣.基于不同隐马尔科夫模型的图像识别方法[J].现代电子技术,2012,35(8):54-56,60.
作者姓名:李楠  姬光荣
作者单位:中国海洋大学信息科学与工程学院,山东青岛,266100
基金项目:国家“863”基金资助项目:有害赤潮生物诊断系统技术研究(2006AA09Z178)
摘    要:为了更详细地研究隐马尔科夫模型在图像识别中的应用,以指纹识别为例,纵向总结了几种基于隐马尔科夫模型的指纹图像识别算法,包括一维隐马尔科夫模型、伪二维隐马尔科夫模型、二维模型及一维模型组。分别从时间复杂度、识别精确度等方面总结出这四种隐马尔科夫模型在图像识别时的优缺点,得出不同待识别图像适合使用的识别模型的结论。

关 键 词:隐马尔科夫模型  E-HMM  图像识别  指纹识别

Image recognition algorithms based on different hidden Markov models
LI Nan , JI Guang-rong.Image recognition algorithms based on different hidden Markov models[J].Modern Electronic Technique,2012,35(8):54-56,60.
Authors:LI Nan  JI Guang-rong
Affiliation:(College of Information Science & Engineering,Ocean University of China,Qingdao 266100,China)
Abstract:For a more detailed study on the applications of hidden Markov models in image recognition,taking fingerprint identification as an example,several fingerprint image recognition algorithms based on HMM(hidden Markov model) are summarized.It includes one-dimensional hidden Markov model,pseudo-two-dimensional hidden Markov model,two-dimensional model and one-dimensional model group.The advantages and disadvantages of the four HMMs in the aspect of image recognition are respectively elaborated in time complexity of recognition accuracy,etc.The classification of the models corresponding to different images under recognition was implemented.
Keywords:hidden Markov model  E-HMM  image recognition  fingerprint recognition
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