摘要
针对低信噪比海情下的小目标检测效果不理想的问题,利用消除趋势波动分析法提取海杂波分形参数,分析了海杂波的功率谱及其熵特征,提出了海杂波背景下的联合特征目标检测方法。在雷达回波中提取海杂波的分形参数和功率谱熵特征组成二维向量,利用凸包训练算法获得纯海杂波判别区域,同样对先验知识下的待测海杂波提取这两个特征参量,以此特征参量所对应的点是否在此判别区域内来判别是否存在目标。利用加拿大IPIX雷达数据,证明了所提算法优于用单个特征差异作为统计量的方法,在相同虚警概率下检测效果明显提高,为雷达目标检测提供了新的检测方案。
For the small target detection effect is not ideal in low signal to clutter sea condition,detrended fluctuation analysis method is used to determine the fractal parameter of sea clutter in this paper,and the sea clutter power spectrum as well as it＇s entropy characteristics is analyzed.At last,a joint target detection method integrating two sea clutter characteristics is proposed.Extract two specific features from sea clutter to compose a two-dimensional vector,utilize convex hull training algorithm to obtain decision region by analyzing pure sea clutter,extract the same two features from the sea clutter with prior knowledge by using the same method,discriminate whether there is a goal by judging whether the corresponding points are in the area or not.The method has been confirmed with the data collected by Master IPIX radar at the east coast of Canada that the algorithm is superior to the contrast algorithms,which provides a new detection scheme for radar target detection.
出处
《现代雷达》
CSCD
北大核心
2014年第10期28-32,共5页
Modern Radar
基金
国家自然科学基金资助项目(61072133)
江苏省产学研联合创新资金计划(BY2013007-02
BY2011112)
江苏省高校科研成果产业化推进项目(JHB2011-15)
江苏省"信息与通信工程"优势学科资助项目
关键词
目标检测
联合特征
凸包
海杂波
object detection
joint characteristic
convex hull
sea clutter
