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

基于SVM的视频中闪光灯的检测方法
引用本文:黄志华,王国乾.基于SVM的视频中闪光灯的检测方法[J].福州大学学报(自然科学版),2009,37(3).
作者姓名:黄志华  王国乾
作者单位:黄志华,HUANG Zhi-hua(福州大学数学与计算机科学学院,福建,福州,350108;厦门大学信息科学与技术学院,福建,厦门,361005);王国乾,WANG Guo-qian(福建省计算中心,福建,福州,350003)  
摘    要:在分析大量实验数据基础上,比较镜头切变、字幕突现或消失、光照突变等与闪光灯具有类似特征的情况后,提出了构造符号特征值的方法.符号特征值能够使得闪光灯与其他类似情况的区别更明显.在符号特征值的基础上提出了构造基于SVM的闪光灯检测器的方法,该检测器在实验中表现良好.

关 键 词:闪光灯检测  符号特征值  检索

Flash detection in video based on SVM
HUANG Zhi-hu,WANG Guo-qian.Flash detection in video based on SVM[J].Journal of Fuzhou University(Natural Science Edition),2009,37(3).
Authors:HUANG Zhi-hu  WANG Guo-qian
Institution:1.College of Mathematics and Computer Science;Fuzhou University;Fuzhou;Fujian 350108;China;2.College of Information Science and Technology;Xiamen University;Xiamen;Fujian 361005;3.Computer Center of Fujian Province;Fujian 350003;China
Abstract:After comparing flash with other situations similar to flash such as cut,caption abrupt appearing or disappearing,and illumination abrupt changing based on analysis of extensive experimental data,we propose a novel method to compute the sign feature,which is proved significantly more efficient than current practices to distinguish flash from other similar situations.A new method is developed to build SVM-based flash detector based on the sign feature and is verified to perform well in experiments.
Keywords:SVM
本文献已被 CNKI 万方数据 等数据库收录!
点击此处可从《福州大学学报(自然科学版)》浏览原始摘要信息
点击此处可从《福州大学学报(自然科学版)》下载免费的PDF全文
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