首页 | 本学科首页   官方微博 | 高级检索  
     

桥区运动船舶的检测和跟踪
引用本文:张文涛,谭毅华,余晋刚,田金文. 桥区运动船舶的检测和跟踪[J]. 计算机与数字工程, 2011, 0(10): 67-71,156
作者姓名:张文涛  谭毅华  余晋刚  田金文
作者单位:华中科技大学多谱信息处理技术国家级重点实验室 武汉 430074
基金项目:国家自然科学基金(编号:50909079);教育部专项基金(编号:109109)资助
摘    要:基于桥区的场景特点,提出一种基于自适应纹理特征的运动船舶目标检测跟踪算法。算法采用基于改进的局部二值模式(LBP)纹理特征取代亮度特征,在此基础上建立可实时更新的LBP背景,以背景模型方法提取船舶运动目标前景,并以简单的近邻方法实现对其跟踪。实验表明该方法通过目标检测与跟踪可实现对船舶运动的监控,算法准确性好,精度高,可为桥梁安全保护提供有效的技术支撑。

关 键 词:运动目标检测  运动目标跟踪  局部二值模式  船舶监控

Detection and Tracking of Moving Ship in Bridge Area
ZhangWentao,Tan Yihua,Yu Jinggang,Tian Jinwen. Detection and Tracking of Moving Ship in Bridge Area[J]. Computer and Digital Engineering, 2011, 0(10): 67-71,156
Authors:ZhangWentao  Tan Yihua  Yu Jinggang  Tian Jinwen
Affiliation:Zhang Wentao Tan Yihua Yu Jinggang Tian Jinwen(National Key Laboratory of Science & Technology-on Multi-Spectral Information Processing, Huazhong University of Science and Technology,Wuhan 430074)
Abstract:A moving ship detection and tracking method based on the adaptive texture features of the bridge area is presented.The background model is established by replacing brightness of LBP texture features to establish the background model,and updated for every coming frame in real time.Foreground areas are extracted as moving ships by adaptive threshold.Finally,the moving ships are tracked using simple nearest neighboring method.Experiments show that it's accurate and precise to detect and track moving ship in the actual video.Therefore,as a crucial technique,it is able to be applied to the bridge protection system.
Keywords:moving target detection  moving target tracking  local binary pattern  ship dynamic monitoring
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