深度学习在水声目标识别中的应用研究

Research on Application of Deep Learning in Underwater Acoustic Target Recognition

摘要:

水声目标识别技术是水声信号处理的重要组成部分，是水声信息获取与水声信息对抗的重要技术支撑。针对水声目标识别时探测数据量大、自动化程度不高、识别效率低下等问题，研究了深度学习在水声目标识别中的应用。首先，介绍了水声目标识别技术的研究现状及当前形势下面临的挑战。然后，对深度学习的网络结构原理及改进型进行了分析，并分别对深度学习在水声声信号识别领域和水声图像信号识别领域的应用现状做了阐述。最后，指出了由于受当前技术条件和水下复杂环境的制约，此方法尚且存在着不足之处。该方法为进一步优化深度学习算法、拓展深度学习技术应用范畴、提升水声目标识别效率提供了参考。

Abstract:

Underwater acoustic target recognition technology is an important part of underwater acoustic signal processing and an important technical support for underwater acoustic information acquisition and underwater acoustic information countermeasures. Aiming at the problems of large amount of detection data，low automation degree and low recognition efficiency in underwater acoustic target recognition，the application of deep learning in underwater acoustic target recognition was studied. First，the research status and the current challenges of underwater acoustic target recognition technology are introduced. Then，the network structure principle and improved form of deep learning are analyzed，and the application statuses of deep learning in underwater acoustic signal recognition and underwater acoustic image signal recognition are explained respectively. Finally，it is pointed out that due to the constraints of current technical conditions and the complex underwater environment，this method still has shortcomings. This mehod provides a reference for further optimization of deep learning algorithms，expansion of the scope of deep learning technology applications，and improvement of underwater acoustic target recognition efficiency.