基于多域多维特征融合的海面小目标检测

Detection of Sea-surface Small Target Based on Multi-domain and Multi-dimensional Feature Fusion

多维特征检测技术是提高海面小目标检测的有效途径。为了进一步提升海面小目标检测性能,本文提出基于多域多维特征融合的检测方法。首先,从时域、频域、时频域、极化域等多域,充分挖掘海杂波和含目标回波的差异性,并将这些差异性表征为多维特征,构建高维特征空间。其次,通过极化域和特征域的多维特征线性融合,将多维特征压缩到3D特征空间中,获得高维度信息的同时减少维度计算代价。然后,结合凸包学习算法获得3D判决区域,实现异常检测。最后,基于IPIX实测数据的实验结果表明:相对现有的极化检测器,提出的检测器具有25%以上的显著性能提升。

Multi-dimensional feature detection technology is an effective way to improve detection performance of sea-surface small targets.A detection method based on multi-domain and multi-dimensional feature fusion is proposed to further improve performance in this paper.First,the differences between sea clutter and target returns are fully explored in time domain,frequency domain,time-frequency domain and polarization domain,which are represented as multi-dimensional features to construct high-dimensional feature space.Second,multi-dimensional features are compressed into 3-dimentional feature space by the linear fusion in polarization domain and feature domain,which can obtain high-dimensional information and reduce dimensional computational cost at the same time.Third,convex hull learning algorithm is used to obtain the 3 D decision region and realize the anomaly detection.Finally,experimental results via IPIX data show that the proposed detector can attain significant performance improvement of more than 25%,relative to the existing polarization detectors.