首页 | 本学科首页   官方微博 | 高级检索  
     

动态场景下的快速目标检测算法
引用本文:齐美彬,汪巍,蒋建国,沈玉亮.动态场景下的快速目标检测算法[J].电子测量与仪器学报,2011,25(9):756-761.
作者姓名:齐美彬  汪巍  蒋建国  沈玉亮
作者单位:1. 合肥工业大学计算机与信息学院,合肥,230009
2. 合肥工业大学计算机与信息学院,合肥230009;安全关键工业测控技术教育部工程研究中心,合肥230009
基金项目:中央高校基本科研业务费专项资金(编号:2011HGZL0001)资助项目
摘    要:提出了一种动态场景下的目标快速检测算法,基于特征点SURF算法,采用旋转参数模型,结合最小二乘法求解全局运动参数进行运动补偿,最后使用帧差法获得运动目标.采用ORSA的方法去除外点的影响;SURF算法不能满足实时性的需要,针对这一点,提出基于运动预测的特征点匹配算法,提高了匹配的速度和准确率;采用基于残差图像块的更新策...

关 键 词:目标检测  运动补偿  SURF  旋转模型

Rapid moving object detection under a dynamic scene
Qi Meibin,Wang Wei,Jiang Jianguo,Shen Yuliang.Rapid moving object detection under a dynamic scene[J].Journal of Electronic Measurement and Instrument,2011,25(9):756-761.
Authors:Qi Meibin  Wang Wei  Jiang Jianguo  Shen Yuliang
Affiliation:Qi Meibin1 Wang Wei1 Jiang Jianguo1,2 Shen Yuliang1 (1.School of Computer and information,Hefei University of Technology,Hefei 230009,China,2.Engineering Research Center of Safety Critical Industrial Measurement and Control Technology,Ministry of Education,Hefei 23009,China)
Abstract:A method for detecting moving object from videos is provided when undergoes a dynamic scene.The method is based on the algorithm of the SURF and the global motion of the image is modeled as rotational parameters,the least square method can precisely compute the motion parameters which then leads to the implementation of motion compensation and the object detection.In this process we using ORSA algorithm to eliminate the outliers which is gen-erated in the process of feature points matching,after that the le...
Keywords:moving object detection  motion compensation  SURF  rotational model  
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