基于组合分形特征的海杂波中目标检测方法  

Target Detection Method in Sea Clutter Based on Combined Fractal Characteristics

作　　者：刘宁波[1,2] 丁昊[1] 田永华 文树梁[2] 关键[1] Liu Ningbo;Ding Hao;Tian Yonghua;Wen Shuliang;Guan Jlan(Information Fusion Institute, Naval Aeronautical and Astronautical University, Yantai 264001, China;No. 23 Inst. of the Second Academy, China Aerospace Science and Industry Corporation, Beijing 100854, China)

机构地区：[1]海军航空工程学院信息融合研究所,山东烟台264001 [2]中国航天科工集团第二研究院二十三所,北京100854

出　　处：《航空兵器》2018年第2期38-42,共5页Aero Weaponry

基　　金：国家自然科学基金项目(61531020;61471382;61501487);中国博士后科学基金项目(2017M620862);航空科学基金项目(20150184003);"泰山学者"建设工程专项经费

摘　　要：仅利用一种分形特征检测海杂波中目标时,在低信杂比时往往难以有效区分。为此,提出一种基于组合分形特征的海杂波中目标检测方法,将检测流程划分为粗检测和精检测两个阶段。当检测单元分形谱维数特征明显时,在粗检测阶段即判决目标存在;当分形谱维数特征不明显时,则以分形模型拟合误差和分维尺度变化量为特征量,利用综合模糊隶属度进行判决,实现目标精检测。经X波段雷达实测数据验证表明,所提检测方法可以较大幅度提升海杂波中弱小目标的检测准确度。Using one of fractal characteristics to detect targets in sea clutter, it is often difficult to distinguish at low signal-to-clutter ratios. To solve this problem, a new method for target detection in sea clutter based on combined fractal characteristics is proposed. The detection process is divided into two sta- ges： rough detection and fine detection. When the fractal spectrum dimension characteristic of the detec- tion unit is obvious, in the rough detection stage, the COXTesponding processing is carried out to achieve the target existence decision. When the fractal spectrum dimension characteristic is not obvious, the frac- tal model fitting error and the fractal dimension change quantity are used as characteristic quantities, and the comprehensive fuzzy membership degree is adopted to crecy out the fine target detection. Verified by X-band radar real data, the results show that the proposed method can greatly enhance the detection accu- racy of weak targets in sea clutter.

关 键 词：目标检测 海杂波 组合分形 谱维数 

分 类 号：TJ760.1[兵器科学与技术—武器系统与运用工程] TN957.51[电子电信—信号与信息处理]