-
针对海杂波背景下海情对小目标检测的严重影响, 本文研究了实测海杂波在分数阶Fourier变换(FRFT)域的分形特征, 分别提出了单、高尺度下的分形检测方法. 由数学定义推得, FRFT 在不同阶数和尺度情况下, 不具有一致的自相似特性, 采用多重分形趋势波动分析法确定分形参数H(q), 分析了海杂波在不同海情、距离和极化条件下的分形特征. 在单尺度基础上结合FRFT的变阶优势, 提出了阶数自适应的小目标检测方法; 高尺度条件下, 比较了不同因素对海杂波FRFT域多重分形参数的影响. 结果表明:海杂波FRFT域可用变换阶数的方法检测到湮没在复杂海情中的小信号, 检测门限多数提高200%以上, 比采用时域信号提高26.3%. H(q) 在负高尺度上具有明显的多重分形特征差异, H(q)-q曲线满足反正切分布, 纯海杂波与含目标数据的拟合幅值比分别大于1.8(HH)和1.4(VV), 为海杂波背景小目标检测提供了判定依据.Because the detection of small target in the background of sea clutter is strongly influenced by the sea condition, this paper studies the fractal property of sea clutter in fractional Fourier transform (FRFT) domain and proposes the fractal detection method in single and high dimensions. The FRFT deduced from mathematical definition is not consistent with the self-similar properties in orders and scales. Multifractal detrended fluctuation analysis (MF-DFA) method is used to determine the fractal parameter H(q) and analyze the fractal property of the sea clutter in different situations, distances, and polarizations. In single dimension, the small target detection method is proposed based on an adaptive order. By comparing different factors of the multifractal parameters, the results show that the transform order method in sea clutter FRFT domain can detect small signals under complicated sea conditions. The detection threshold mostly increases above 200%, which is 26.3% higher than the method of time domain signal. H(q) has an obvious multifractal difference on high negative scale, the H(q)-q curve satisfies the arctangent distribution. The fitting amplitude ratios of pure sea clutter and target data are greater than 1.8 (HH) and 1.4 (VV), which provide the basis for the small target detection in sea clutter background.
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] -
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28]
计量
- 文章访问数:6576
- PDF下载量:264
- 被引次数:0