论文部分内容阅读
提出了一种基于虚拟多传感器融合技术的红外目标识别方法.文中利用傅里叶描述器提取目标形状的边缘特征以及辐射特性的六个特征量,采用多个人工神经网络对来自单一传感器的目标利用不同特征分别识别,再利用D-S证据推理将各个网络的识别结果进行决策级融合.仿真实验结果表明,该方法提高了识别率和识别结果的可靠性.
An infrared target recognition method based on fictitious multisensor fusion technology is proposed.In this paper, the Fourier descriptor is used to extract the edge features of the target shape and the six characteristic parameters of the radiation characteristics, and a number of artificial neural networks The recognition results of each network are merged at decision level by using DS evidence reasoning.The simulation results show that this method improves the recognition rate and the reliability of the recognition results.