基于D-S证据理论的近岸舰船目标识别

在深入分析和提取近岸舰船目标红外特性和微波特性的基础之上,提出一种基于D-S证据理论的红外图像和微波特性融合方法来实现近岸舰船目标的识别.一方面,对红外图像进行小波去噪、边缘提取、海天线识别与对消等红外预处理和特征提取后,可以得到若干个待选目标;另一方面,对目标回波进行衰减、低噪声放大、滤波和线性放大等处理后,可以得到信号幅度大于一定阈值若干个待选目标.最后,通过对红外和微波得到的待选目标运用D-s证据理论的融合规则进行数据融合,得到最终识别结果.算法易于硬件实现.实验表明,该方法能达到较好的识别效果.

Alongshore Ship Detection Based on D-S Evidence Theory

Based on analyzing and extracting the infrared and microwave features of alongshore ships,an approach is presented to recognize the ships by fusing the data of infrared and microwave based on D-S Evidence Theory.Through infrared preprocessing and feature extracting to the infrared images,such as wavelet-based denoising,edge extracting,sea-sky line identifying etc,some possible targets can be obtained.On the other hand,through target echo processing,such as attenuating,lower-level noise amplifying,filtering,...