海杂波的多重分形判定及广义维数谱自动提取

Judgement of Multifractal and Auto-computing of Generalized Dimension Spectrum Based on Sea Clutter

摘要:

海杂波中的微弱目标检测历来是一个难题,传统的检测方法一直不能得到较好的检测效果。多重分形理论的引入为海杂波中微弱目标检测提供了一个很好的途径。文章提出了多重分形无标度区间自动识别的改进算法,既保留了该算法的客观性,又降低了它的保守程度,使估计的区间更贴近真实无标度区间,同时实现了广义维数的自动计算。实验证明,该方法优于原方法,结果更为精确。

Abstract:

Weak targets detection in sea clutter is always a difficult problem, and the traditional methods cannot get better effect. However, the multifractal theory offers a method to resolve it. The method of the auto-recognizing of multifractal scale-invariant space interval was put forward. The improved method both reduces the insufficiency of the original method and keeps its impersonality. So the space interval from the improved method is much close to the real interval. At the same time, the improved method can compute the generalized dimension automatically. Experiments showed that the new method was better than the old and could offer a more accurate result.