首页 | 本学科首页   官方微博 | 高级检索  
     

风力机叶片的图像跟踪与识别算法研究
引用本文:沈继忱,刘志杰,邸建铭,赵世荣.风力机叶片的图像跟踪与识别算法研究[J].黑龙江电力,2012,34(3):167-170.
作者姓名:沈继忱  刘志杰  邸建铭  赵世荣
作者单位:东北电力大学自动化工程学院,吉林 吉林,132012
摘    要:为及时发现、处理风力机叶片事故,提高风力发电的安全性和可靠性,针对中国现行风力发电的特点,采用图像跟踪技术,借助Adaboost算法对风力机叶片运行状态进行跟踪与识别研究,并提出了Adaboost算法缩短训练耗时间的改进方法。结果表明,改进后的算法可以减少训练分类器的时间,使跟踪、识别的实时性与准确性更为理想。

关 键 词:叶片检测  模式识别  分类器  Adaboost算法

Study on the image tracking and recognition algorithm of fan blade
SHEN Jichen , LIU Zhijie , DI Jianming , ZHAO Shirong.Study on the image tracking and recognition algorithm of fan blade[J].Heilongjiang Electric Power,2012,34(3):167-170.
Authors:SHEN Jichen  LIU Zhijie  DI Jianming  ZHAO Shirong
Affiliation:(School of Automation Engineering of Northeast Dianli University, Jilin 132012, China)
Abstract:In order to find and deal with the fan blade accident in time and enhance the security and reliability of wind power generation, this paper studies, on the basis of the characteristics of wind power generation in our country, the tracking and recognition of the running state of fan blade by image tracking and with the help of Adaboost algorithm, and proposes measures to shorten the training time by Adaboost algorithm. The results show that the improved algorithm is able to shorten training time of classifier and achieve better real time and accuracy of recognition.
Keywords:blade detection  mode recognition  classifier  Adaboost algorithm
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