中文摘要：

水下沉底或掩埋目标识别受海底混响干扰严重，并且目标回波与混响在时一频域上均存在混叠，增大了对目标回波与混响的分离难度。根据目标回波与混响的产生机理差异，将图像形态学与时频分析相结合的盲分离算法用于目标回波与混响的分离。推导了目标几何声散射成分与混响在Wigner—Ville时频面的形态特征表达式，利用图像形态学滤波去除Wigner—Ville时频面中的混响及刚性亮点之间交叉项，在时频形态特征域获取解混矩阵，实现了目标回波和混响的分离。仿真与实验数据处理结果表明，结合图像形态学的时频域盲分离算法提高了目标回波信号的信混比。

英文摘要：

Underwater bottom or buried target recognition is seriously influenced by reverberation, and the target echo is aliased with the reverberation in time-frequency domain, making the separation of target echo and reverberation difficult With the different generating mechanisms between the target echo and reverberation, a time-frequency blind source separation method combined with the image morphology was used to separate the target echo and the reverber- ation. The morphological characteristics of the target geometrical acoustic scattering components and the reverberation in Wigner-Ville distribution are derived Morphological filter was used to remove the reverberation and the cross-terms between the geometrical acoustic scattering components in the Wigner-Ville distribution, then obtained an unmixed ma- trix in the time-frequency morphological domain, so target echo and reverberation separation was achieved. Simulation and experimental data processing results showed that the proposed algorithm can improve the signal to reverberation ratio.