论文部分内容阅读
为了提高机载传感器目标识别系统的性能,提出了利用机载雷达和红外成像传感器数据融合的智能目标识别算法。对红外成像传感器,采用了基于小波矩特征和BP神经网络的目标识别算法,首先提取目标图像的小波矩特征并进行特征选择,然后通过BP神经网络对目标图像进行识别;对雷达传感器,提出了利用模糊推理的目标识别方法,首先选取适当的雷达特征,然后通过模糊推理进行识别;从雷达和红外传感器识别算法分别得到待识别目标所属类别的基本概率分配函数,用D-S证据组合规则将两个基本概率分配函数组合,最终实现了机载雷达和红外传感器的数据融合。仿真结果表明:融合后的识别效果优于单个雷达或红外传感器的识别效果。
In order to improve the performance of airborne sensor target recognition system, an intelligent target recognition algorithm using airborne radar and infrared imaging sensor data fusion is proposed. For the infrared imaging sensor, a target recognition algorithm based on wavelet moment feature and BP neural network is adopted. Firstly, the wavelet moment feature of the target image is extracted and the feature is selected. Then the target image is identified by BP neural network. For the radar sensor, Firstly, the proper radar features are selected and then identified by fuzzy reasoning. The basic probability distribution functions of the categories to be identified are obtained respectively from the radar and infrared sensor recognition algorithms, and two Basic probability distribution function combination, finally realized the airborne radar and infrared sensor data fusion. Simulation results show that the recognition effect after fusion is better than that of single radar or infrared sensor.