在图像制导导弹中,射手经常参与识别、捕获目标以及控制导弹跟踪、攻击目标。为评价一个射手,通过增加输入和输出论域的模糊子集数,对普通的射手模糊神经网络模型进行了改进。基于改进的射手模糊神经网络模型,提出了两种评价射手的方法。仿真结果表明:两种方法都能够有效地比较和评价射手,为选择和训练射手提供了依据。 In image-guided missiles, shooters are often involved in identifying, capturing targets and controlling missile tracking to attack targets. To evaluate a shooter, an improved normal shooter fuzzy neural network model is improved by increasing the number of fuzzy subsets in input and output domains. Based on the improved shooter fuzzy neural network model, two methods for evaluating shooter are proposed. Simulation results show that both methods can effectively compare and evaluate shooters, which provides a basis for selecting and training shooters.