摘要
在数据融合的基础上,以红外／毫米波双模传感器的智能融合结构为模型,将模糊神经网络与D-S证据理论相结合,提出了一种新的目标识别方法。该算法根据红外／毫米波传感器的性能及工作范围,构造模糊变量作为神经网络的输入,根据神经网络的不同输出判别目标的真伪,并利用D-S证据理论进行目标身份识别。仿真结果证明了该算法的可行性。
On the basis of data fusion, a new method of target recognition was proposed by using the structure of infrared/ millimeter waves sensing fusion, and combining neuro-fuzzy network and D-S evidence theory. Based on the characteristic of infrared and millimeter waves, this method constructs fuzzy variable as the input of neural network. Using the D-S evidence reasoning, target is distinguished with the output of neural network. The experiment results show that the proposed algorithm achieves a good segmentation quality.
出处
《光学与光电技术》
2006年第4期62-64,共3页
Optics & Optoelectronic Technology
